Arxiv:1912.00017V1 [Physics.Atom-Ph] 29 Nov 2019

Theoretical Atto-nano Physics

Marcelo F. Ciappina1 and Maciej Lewenstein2, 3 1Institute of Physics of the ASCR, ELI-Beamlines, Na Slovance 2, 182 21 Prague, Czech Republic 2ICFO - Institut de Ciencies Fotoniques, The Barcelona Institute of Science and Technology, Av. Carl Friedrich Gauss 3, 08860 Castelldefels (Barcelona), Spain 3ICREA - InstitucióCatalana de Recerca i Estudis Avan¸cats,Lluis Companys 23, 08010 Barcelona, Spain (Dated: December 3, 2019) Two emerging areas of research, attosecond and nanoscale physics, have recently started to merge. Attosecond physics deals with phenomena occurring when ultrashort laser pulses, with duration on the femto- and sub-femtosecond time scales, interact with atoms, molecules or solids. The laser- induced electron dynamics occurs natively on a timescale down to a few hundred or even tens of attoseconds (1 attosecond=1 as=10−18 s), which is of the order of the optical field cycle. For com- parison, the revolution of an electron on a 1s orbital of a hydrogen atom is ∼ 152 as. On the other hand, the second topic involves the manipulation and engineering of mesoscopic systems, such as solids, metals and dielectrics, with nanometric precision. Although nano-engineering is a vast and well-established research field on its own, the combination with intense laser physics is relatively recent. We present a comprehensive theoretical overview of the tools to tackle and understand the physics that takes place when short and intense laser pulses interact with nanosystems, such as metallic and dielectric nanostructures. In particular we elucidate how the spatially inhomogeneous laser induced fields at a nanometer scale modify the laser-driven electron dynamics. Consequently, this has important impact on pivotal processes such as above-threshold ionization and high-order harmonic generation. The deep understanding of the coupled dynamics between these spatially inhomogeneous fields and matter configures a promising way to new avenues of research and applications. Thanks to the maturity that attosecond physics has reached, together with the tremendous advance in material engineering and manipulation techniques, the age of atto-nano physics has begun, but it is still in an incipient stage. This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in the 21st Century Nanoscience A Handbook: Nanophysics Sourcebook (Volume One), copyright c CRC Press, Taylor & Francis Group after peer review. The final edited and published work is available at https://doi.org/10.1201/9780367333003

I. INTRODUCTION Simultaneously, bulk matter samples have been down- scaled in size to nanometric dimensions, opening the door to study light-matter interaction in a completely new This chapter deals with an embryonic field of atomic, arena. When a strong laser interacts, for instance, with a molecular, and optical physics: atto-nanophysics. It is an metallic structure, it can couple with the plasmon modes area that combines the traditional and already very ma- inducing the ones corresponding to collective oscillations ture attosecond physics with the equally well developed of free charges (electrons). These free charges, driven by nanophysics. We start our contribution by just present- the field, generate ’nanospots’, of few nanometers size, of ing the general motivations and description of this new highly enhanced near-fields, which exhibit unique tem- area, restricting ourselves to very basic and well known poral and spatial properties. The near-fields, in turn, in- concepts and references. duce appreciable changes in the local field strength at a Attosecond physics has traditionally focused on atomic scale of the order of tenths of nanometers, and in this way and small molecular targets [1, 2]. For such systems the they play an important role modifying the field-induced electron, once it is ionized by the electric field of the laser electron dynamics. In other words, in this regime, the pulse, moves in a region that is small compared to the spatial scale on which the electron dynamics takes place wavelength of the driving laser. Hence, the spatial de- is of the same order as the field variations. Moreover, the pendence of the laser field can be safely neglected. In near-fields change on a sub-cycle timescale as the free arXiv:1912.00017v1 [physics.atom-ph] 29 Nov 2019 the presence of such so-called ’spatially homogeneous’ charges respond almost instantaneously to the driving laser fields the time-dependent processes occurring on laser. Consequently, we face an unprecedented scenario: the attosecond time scale have been extensively inves- the possibility to study and manipulate strong-field in- tigated [3, 4]. This subject has came to age based upon duced processes by rapidly changing fields, which are not well-established theoretical developments and the under- spatially homogeneous. In the following subsections we standing of numerous nonlinear phenomena (cf. [5–7]), present a description of these strong field processes, joint as well as the tremendous advances in experimental laser with the theoretical tools particularly modified to tackle techniques. For instance, nowadays experimental mea- them. The emergent field of attosecond physics at the surements with attosecond precision are routinely per- nanoscale marries very fast attosecond processes (1 as formed in several facilities around the world. = 10−18 s), with very small nanometric spatial scales (1 2 nm = 10−9 m), bringing a unique and sometimes unex- cess, i.e. the number of photon necessary to overpass the pected perspective on important underlying strong field ionization potential. phenomena. Many experiments take place in an intermediate or cross-over region, defined by γ ∼ 1 [14]. Another way to interpret γ is to note that γ =√ τT /τL, where τT is the A. Strong field phenomena driven by spatially 2Ip Keldysh time (defined as τT = ) and τL is the laser homogeneous fields E0 period. Hence γ serves as a measure of non-adiabaticity by comparing the response time of the electron wavefunc- A particular way of initiating electronic dynamics in tion to the period of the laser field. atoms, molecules and recently solid materials, is to ex- When laser intensities approach 1013 ∼ 1014 W cm−2, pose these systems to an intense and coherent electro- the usual perturbative scaling observed in the multipho- magnetic radiation. A variety of widely studied and im- ton regime (γ 1) does not hold, and the emission portant phenomena, which we simply list and shortly process becomes dominated by tunnelling (γ < 1). In describe here, result from this interaction. In order to this regime a strong laser field bends the binding poten- first put the relevant laser parameters into context, it tial of the atom creating a penetrable potential barrier. is useful to compare them with an atomic reference. In The ionization process is governed thus by electrons tun- the present framework, laser fields are considered intense nelling through this potential barrier, and subsequently when their strength is not much smaller or even compa- interacting ”classically” with the strong laser field far rable to the Coulomb field experienced by an atomic elec- from the parent ion [15–17]. tron. The Coulomb field in an hydrogen atom is approxi- This concept of tunnel ionization underpins many im- 9 −1 −1 mately 5×10 V cm (≈ 514 V nm ), corresponding to portant theoretical advances, which have received crys- 16 −2 an equivalent intensity of 3.51 × 10 W cm – this last tal clear experimental confirmation with the development value indeed defines the atomic unit of intensity. With of intense ultra-short lasers and attosecond sources over regard to time scales, we note that in the Bohr model of the past two decades. On a fundamental level, theoret- hydrogen atom, the electron takes about 150 as to orbit ical and experimental progress opened the door to the around the proton, defining thus the characteristic time study of basic atomic and molecular processes on the at- for electron dynamics inside atoms and molecules [8]. Fi- tosecond time scale. On a practical level, this led to the nally, the relevant laser sources are typically in the near- development of attosecond high frequency extreme ul- infrared (NIR) regime, and hence laser frequencies are traviolet (XUV) and X-ray sources, which promise many much below the ionisation atomic or molecular poten- important applications, such fine control of atomic and tial. In particular, an 800 nm source corresponds to a molecular reactions among others. The very fact that we photon energy of 1.55 eV (0.057 au), which is much be- deal here with sources that produce pulses of attosecond low the ionisation potential of hydrogen, given by 1/2 au duration is remarkable. Attosecond XUV pulses allow, (13.6 eV). At the same time, laser intensities are in the in principle, to capture all processes underlying struc- 13 15 −2 10 − 10 W cm range: high enough to ionize a no- tural dynamics and chemical reactions, including elec- ticeable fraction of the sample, but low enough to avoid tronic motion coupled to nuclear dynamics. They allow space charge effects and full depletion of the ground state. also to address basic unresolved and controversial ques- While the physics of interactions of atoms and tions in quantum mechanics, such as, for instance, the molecules with intense laser pulses is quite complex, duration of the strong field ionization process or the tun- much can be learnt using theoretical tools developed nelling time [14, 18]. over the past decades, starting with the original work by Among the variety of phenomena which take place Keldysh in the 1960’s [9–13]. According to the Keldysh when atomic systems are driven by coherent and in- theory, an electron can be freed from an atomic or molec- tense electromagnetic radiation, the most notable exam- ular core either via tunnel or multiphoton ionization. ples are high-order harmonic generation (HHG), above- These two regimes are characterized by the Keldysh pa- threshold ionization (ATI) and multiple sequential or rameter: non-sequential ionization. All these processes present similarities and differences, which we detail briefly be- p s low [6, 7, 19] 2Ip Ip HHG takes place whenever an atom or molecule in- γ = ω0 = , (1) E0 2Up teracts with an intense laser field of frequency ω0, pro- ducing radiation of higher multiples of the fundamental where Ip is the ionization potential, Up is the pondero- frequency Kω0, where in the simplest case of rotation- 2 2 motive energy, defined as Up = E0 /4ω0 where E0 is the ally symmetric target K is an odd integer. HHG spec- peak laser electric field and ω0 the laser carrier frequency. tra present very distinct characteristics: there is a sharp The adiabatic tunnelling regime is then characterized by decline in conversion efficiency followed by a plateau in γ 1, whereas the multiphoton ionization regime by which the harmonic intensity hardly varies with the har- γ 1. In the multiphoton regime ionisation rates scale monic order K, and eventually an abrupt cutoff. For an as laser intensity IN , where N is the order of the pro- inversion symmetric medium (such as all atoms and some 3 molecules), only odd harmonics of the driving field have tron. This picture changes dramatically when few-cycle been observed because of dipole selection rules and the pulses are used to drive the media and the ATI energy central symmetric character of the potential formed by spectra become much richer structurally speaking [28]. the laser pulse and the atomic field. The discovery of this In this case, we can clearly distinguish two different plateau region in HHG has made the generation of coher- regions, corresponding to the direct and rescattered elec- ent XUV radiation using table-top lasers feasible. The trons, respectively. The low energy region, given by above mentioned features characterize a highly nonlinear Ek . 2Up, corresponds to direct electrons or electrons process [20]. Furthermore, HHG spectroscopy (i.e. the which never come back to the vicinity of the parent atom. measurement and interpretation of the HHG emission On the other hand, the high energy part of the ATI spec- from a sample) has been widely applied to studying the trum 2Up . Ek . 10Up is dominated by the rescattered ultrafast dynamics of molecules interacting with strong electrons, i.e. the electrons that reach the detector after laser fields (see, e.g. [21]). being rescattered by the remaining ion-core [29]. The Conceptually, HHG is easily understood using the latter are strongly influenced by the absolute phase of a three-step model [15, 17, 22–24]: (i) tunnel ionization due few-cycle pulse and as a consequence they are used rou- to the intense and low frequency laser field; (ii) acceler- tinely for laser pulse characterization [30]. These two en- ation of the free electron by the laser electric field, and ergy limits for both the direct and rescattered electrons, (iii) re-collision with the parent atom or molecular ion. i.e. 2Up and 10Up can be easily obtained invoking purely The kinetic energy gained by the electron in its travel, classical arguments [28, 31, 32]. under the presence of the laser oscillatory electric field, Most of the ATI and HHG experiments use as an inter- is converted into a high energy photon and can be easily acting media multielectronic atoms and molecules, and calculated starting from semiclassical assumptions. recently condensed and bulk matter. Nevertheless, one HHG has received special attention because it config- often assumes that only one valence electron is active and ures the workhorse for the creation of attosecond pulses hence determines all the significant features of the strong and, simultaneously, it exemplifies a special challenge field laser-matter interaction. The first observations of from a theoretical point of view due to the complex inter- two-electron effects in ionization by strong laser pulses twining between the Coulomb and external laser fields. go back to the famous Anne L’Huillier’s ‘knee’ [33]. This Additionally, HHG is a promising way to provide a co- paper and later the influential Paul Corkum’s work [15] herent table-top sized short wavelength light sources in stimulated the discussion about sequential versus non- the XUV and soft x-ray regions of the spectrum. Non- sequential ionization, and about a specific mechanism linear atom-electron dynamics triggered by focusing in- of the latter (shake-off, rescattering, etc.). In the last tense laser pulses onto noble gases generates broadband twenty years, and more recently as well, there has been a high photons whose energy reaches the soft X-ray re- growing interest in electron correlations, both in single- gion. This nonlinear phenomenon requires laser inten- and multi-electron ionization regimes, corresponding to sities in the range of 1014 W cm−2, routinely available lower and higher intensities, respectively (cf. [34–36]). from Ti:sapphire femtosecond laser amplifiers [25]. One prominent example where electron correlation Another widely studied phenomenon is ATI. In fact, plays an instrumental role is the so-called non-sequential and from an historical viewpoint, it was the first one to double ionization (NSDI) [34]. It stands in contrast to be considered as a strong nonperturbative laser-matter sequential double (or multiple) ionization, i.e. when the interaction process [26, 27]. Conceptually, ATI is simi- process undergoes a sequence of single ionization events, lar to HHG, except the electron does not recombine with with no correlation between them. NSDI has attracted the parent atom in the step (iii), but rather it is acceler- considerable interest, since it gives direct experimental ated away by the laser field, eventually being registered access to electron-electron correlation – something that at the detector. Hence, ATI is a much more likely process is famously difficult to analyse both analytically and nu- than HHG, although the latter has opened a venue for a merically (for a recent review see, e.g. [37]). larger set of applications and technological developments. Nevertheless, ATI is an essential tool for laser pulse characterization, in particular in the few-cycle pulses regime. B. Introduction to atto-nano physics Unlike in HHG, where macroscopic effects, such as phase matching, often have to be incorporated to reliably repro- The interaction of ultra-short strong laser pulses with duce the experiment, single atom simulations are gener- extended systems has recently received much attention ally enough for ATI modeling. and led to an advance in our understanding of the at- In an ordinary ATI experiment, the energy and/or an- tosecond to few-femtosecond electronic and nuclear dy- gular distribution of photoelectrons is measured. The namics. For instance, the interaction of clusters with ATI spectrum in energy presents a series of peaks given strong ultrafast laser fields has long been known to lead by the formula Ep = (m + s)ω0 − Ip, where m is the to the formation of nanoplasmas in which there is a minimum number of laser photons needed to exceed the high degree of charge localisation and ultrafast dynamics, atomic binding energy Ip and s is commonly called the with the emission of energetic (multiple keV) electrons number of ‘above-threshold’ photons carried by the elec- and highly charged -up to Xe40+- ions with high energy 4

(a) (MeV scale) [38–42]. Most recently use of short pulses interaction region (∼ 10 fs) has succeeded in isolating the electron dynam- [µm] ics from the longer timescale ion dynamics (which are Detector essentially frozen) revealing a higher degree of fragmen- e– tation anisotropy in both electrons and ions compared

xuv to the isotropic distributions found from longer pulses input pulse (∼ 100 fs) [43]. 10 13 —1014 W/m 2 gas jet Detector Likewise, interactions of intense lasers with nanoparticles, such as micron scale liquid droplets, leads to hot plasma formation. An important role is found for enhanced local fields on the surface of these droplets driving (b) enhanced laser Detector this interaction via “field hotspots” [44–49]. field Furthermore, studies of driving bound and free charges e– in larger molecules, e.g. collective electron dynamics xuv in fullerenes [50], and in graphene-like structures [51], Detector proton migration in hydrocarbon molecules [52], and input pulse gas atoms 10 10 —1011 W/m 2 interaction region charge migration in proteins [53, 54] could be included [nm] in this group. In turn, laser-driven broad-band electron wavepackets have been used for static and dynamic metal nanostructure diffraction imaging of molecules [55–57], obtaining struc- tural information with sub-nanometer resolution. FIG. 1: Sketch of conventional (a) and Tailored ultra-short and intense fields have also been plasmonic-enhanced (b) strong field processes. used to drive electron dynamics and electron or photon emission from (nanostructured) solids (for a recent compilation see e.g. [58]). The progress seen in recent this paper definitively stimulated a constant interest in years has been largely driven by advances in experimen- the plasmonic-enhanced HHG and ATI [77–82]. tal and engineering techniques (both in laser technology Within the conventional assumption, both the laser and in nanofabrication). Among the remarkable achieve- electric field, E(r, t), and the corresponding vector po- ments in just the latest years are the demonstration of tential, A(r, t), are spatially homogeneous in the region driving electron currents and switching the conductiv- where the electron moves and only their time dependence ity of dielectrics with ultrashort pulses [59, 60], control- is considered, i.e. E(r, t) = E(t) and A(r, t) = A(t). ling the light-induced electron emission from nanopar- This is a authentic assumption considering the usual elec- 2 ticles [61, 62] and nanotips [63–65], and the sub-cycle tron excursion (estimated classically using α = E0/ω0) driven photon emission from solids [66–69]. Furthermore, is bounded roughly by a few nanometers in the NIR, for the intrinsic electron propagation and photoemission pro- typical laser intensities, and several tens of nanometers cesses have been investigated on their natural, attosecond 2 for mid-infrared (MIR) sources (note that α ∝ λ0, where√ timescales [70–74]. λ0 is the wavelength of the driving laser and E0 = I, A key feature of light-nanostructure interaction is the where I is the laser intensity) [25]. Hence, electron ex- enhancement (amplification) of the electric near-field by cursion is very small relative to the spatial variation of several orders of magnitude, and its local confinement the field in the absence of local (or nanoplasmonic) field on a sub-wavelength scale [75]. From a theoretical view- enhancement (see Fig. 1(a)). On the contrary, the fields point, this field localisation presents a unique challenge: generated using surface plasmons are spatially dependent we have at our disposal strong fields that change on on a nanometric region (cf. Fig. 1(b)). As a conse- a comparable spatial scale of the oscillatory electron quence, all the standard theoretical tools in the strong dynamics that are initiated by those same fields. As field ionization toolbox (ranging from purely classical to will be shown throughout this contribution, this singular frequently used semiclassical and complete quantum me- property entails profound consequences in the underly- chanical approaches) have to be re-pondered. In this ing physics of the conventional strong field phenomena. contribution, we will therefore concentrate our efforts on In particular, it defies one of the main assumptions that how the most important and basic processes in strong modelling of strong-field interactions is based upon: the field physics, such as HHG and ATI, are modified in spatial homogeneity of laser fields in the volume of the this new setting of strong field ultrafast phenomena on electronic dynamics under scrutiny. a nano-scale. Additionally, we discuss how the conven- Interestingly, an exponential growing attention in tional theoretical tools have to be modified in order to be strong field phenomena induced by plasmonic-enhanced suitable for this new scenario. Note that the strong field fields was triggered by the questionable work of Kim et phenomena driven by plasmonic fields could be treated al. [76]. These authors claimed to have been observed theoretically within a particular flavour of a non-dipole efficient HHG from bow-tie metallic nanostructures. Al- approximation, but neglecting completely magnetic ef- though the interpretation of the outcomes was incorrect, fects. 5

II. THEORETICAL APPROACHES nanosystem under study and the input laser pulse characteristics (see e.g. [84]). The electric field retrieved In the next subsections we describe the theoretical ap- numerically is then approximated using a power series PN i proaches we have developed to tackle strong field pro- h(x) = i=1 bix , where the coefficients bi are obtained cesses driven by spatially inhomogeneous laser fields. We by fitting the real electric field that results from a finite put particular emphasis on HHG and ATI. element simulation. Furthermore, in the region relevant for the strong field physics and electron dynamics and in the range of the parameters we are considering, the A. Quantum approaches electric field can be indeed approximated by its linear dependence. The 1D-TDSE can be solved numerically by using the The dynamics of a single active atomic electron in a Crank-Nicolson scheme in order to obtain the time prop- strong laser field takes place along the polarization direc- agated electronic wavefunction Ψ(x, t). Once Ψ(x, t) is tion of the field, when linearly polarized laser pulses drive found, we can compute the harmonic spectrum D(ω) by the system. It is then perfectly legitimate to model the Fourier transforming the dipole acceleration a(t) of the HHG and ATI in a 1D spatial dimension by solving the active electron. That is, time-dependent Schrödinger equation (1D-TDSE) [83]: Z ∞ 2 1 1 −iωt ∂Ψ(x, t) D(ω) = dte a(t) , (6) i = H(t)Ψ(x, t) (2) T ω2 ∂t p −∞ 1 ∂2 where Tp is the total duration of the laser pulse. a(t) can = − + V (x) + V (x, t) Ψ(x, t), 2 ∂x2 a l be obtained by using the commutator relation d2hxi where in order to model an atom in 1D, it is common to a(t) = = −hΨ(t)| [H(t), [H(t), x]] |Ψ(t)i, (7) dt2 use soft core potentials, which are of the form: where H(t) is the Hamiltonian specified in Eq. (2). 1 One of the main advantages of the 1D-TDSE is that Va(x) = −√ , (3) x2 + b2 we are able to include any functional form for the spatial variation of the plasmonic field. For instance, we where the parameter b allows us to modify the ionization have implemented linear [83] and real (parabolic) plas- potential Ip of the ground state, fixing it as close as pos- monic fields [84], as well as near-fields with exponential sible to the value of the atom under consideration. We decay (evanescent fields) [88] and gaussian-like bounded consider the field to be linearly polarized along the x-axis spatially fields [87]. and modify the interaction term Vl(x, t) in order to treat In order to calculate ATI-related observables, we use spatially nonhomogeneous fields, while maintaining the the same one-dimensional time-dependent Schrödinger dipole character. Consequently we write equation (1D-TDSE) employed for the computation of HHG [Eq. (2)]. In 1D we are only able to compute the so- Vl(x, t) = −E(x, t) x (4) called energy-resolved photoelectron spectra P (E), i.e. a quantity proportional to the probability to find electrons where E(x, t) is the laser electric field defined as with a particular energy E. In order to do so, we use the window function technique developed by Schafer [89, 90]. E(x, t) = E f(t) (1 + εh(x)) sin(ω t + φ). (5) 0 0 This tool has been widely used, both to calculate angle- resolved and energy-resolved photoelectron spectra [91] In Eq. (5), E0, ω0 and φ are the peak amplitude, the frequency of the laser pulse and the carrier-envelope phase and it represents a step forward with respect to the usual (CEP), respectively. We refer to sin(cos)-like laser pulses projection methods. where φ = 0 (φ = π/2). The pulse envelope is given by An extension of the above described approach is to f(t) and ε is a small parameter that characterizes the in- solve the TDSE in its full dimensionality and to include homogeneity strength. The function h(x) represents the in the laser-electron potential the spatial variation of the functional form of the spatial nonhomogeneous field and, laser electric field. For only one active electron we need to in principle, could take any form and be supported by the deal with 3 spatial dimensions and, due to the cylindrical symmetry of the problem, we are able to separate the numerical algorithm (for details see e.g. [83, 84]). Most m of the approaches use the simplest form for h(x), i.e. the electronic wavefunction in spherical harmonics, Yl and linear term: h(x) = x. This choice is motivated by pre- consider only terms with m = 0 (see below). vious investigations [83, 85, 86], but nothing prevents to In particular, the 3D-TDSE in the length gauge can be use more general functional forms for h(x) [87]. written: ∂Ψ(r, t) The actual spatial dependence of the enhanced near- i = HΨ(r, t) field in the surrounding of a metal nanostructure can ∂t be obtained by solving the Maxwell equations incorpo- ∇2 = − + V (r) + V (r, t) Ψ(r, t), (8) rating both the geometry and material properties of the 2 SAE l 6 where VSAE(r) is the atomic potential in the single ac- we obtain a set of coupled differential equations for each tive electron (SAE) approximation and Vl(r, t) the laser- of the radial functions Φl(r, t): electron coupling (see below). The time-dependent elec- 2 tronic wave function Ψ(r, t), can be expanded in terms ∂Φl 1 ∂ l(l + 1) 1 i = − + − Φl of spherical harmonics as: ∂t 2 ∂r2 2r2 2 2 2 2 Ψ(r, t) = Ψ(r, θ, φ, t) +εr E(t) cl + cl−1 Φl L−1 l +rE(t)(cl−1Φl−1 + clΦl+1) X X Φlm(r, t) ≈ Y m(θ, φ) (9) +εr2E(t)(c c Φ + c c Φ ) . (14) r l l−2 l−1 l−2 l l+1 l+2 l=0 m=−l Equation (14) is solved using the Crank-Nicolson algo- where the number of partial waves depends on each spe- rithm considering the additional term, i.e. Eq. (12) due cific case. Here, in order to assure the numerical conver- to the spatial inhomogeneity. As can be observed, the gence, we have used up to L ≈ 250 in the most extreme 14 2 degree of complexity will increase substantially when a case (I ∼ 5 × 10 W/cm ). In addition, due to the fact more complex functional form for the spatial inhomoge- that the plasmonic field is linearly polarized, the mag- neous laser electric field is used. For instance, the incor- netic quantum number is conserved and consequently in poration of only a linear term couples the angular mo- the following we can consider only m = 0 in Eq. (9). menta l, l ± 1, l ± 2, instead of l, l ± 1, as in the case of This property considerably reduces the complexity of the conventional (spatial homogeneous) laser fields. problem. In here, we consider z as a polarization axis As was already mentioned, typically several hundreds and we take into account that the spatial variation of the of angular momenta l should to be considered and we electric field is linear with respect to the position. As a could recognize the time evolution of each of them as result, the coupling Vl(r, t) between the atomic electron a 1D problem. We use a Crank-Nicolson method im- and the electromagnetic radiation reads plemented on a splitting of the time-evolution operator Z r that preserves the norm of the wave function for the time 0 0 Vl(r, t) = dr · E(r , t) = E0z(1 + εz)f(t) sin(ω0t + φ) propagation, similar to the 1D-TDSE case. The har- (10) monic spectrum D(ω) is then computed in the same was as in the 1D case, but now using the 3D electronic wave- where E0, ω0 and φ are the laser electric field amplitude, the central frequency and the CEP, respectively. functions Ψ(r, t). As in previous investigations, the parameter ε defines We have also made studies on helium because a ma- the ‘strength’ of the inhomogeneity and has units of in- jority of experiments in HHG are carried out in noble verse length (see also [83, 85, 86]). For modeling short gases. Nonetheless, other atoms could be easily imple- laser pulses in Eq. (10), we use a sin-squared envelope mented by choosing the appropriate atomic model potential V (r). After time propagation of the electronic 2 ω0t SAE f(t) of the form f(t) = sin , where np is the total 2np wavefunction, the HHG spectra can be computed in an number of optical cycles. As a result, the total duration analogous way as in the case of the 1D-TDSE. Due to the of the laser pulse will be Tp = npτL where τL = 2π/ω0 complexity of the problem, only simulations with nonho- is the laser period. We focus our analysis on a hydrogen mogeneous fields with linear spatial variations along the atom, i.e. VSAE(r) = −1/r in Eq. (8), and we also as- laser polarization in the 3D-TDSE have been studied. sume that before switch on the laser (t = −∞) the target This, however, is enough to confirm that even a small spa- atom is in its ground state (1s), whose analytic form can tial inhomogeneity significantly modifies the HHG spec- be found in a standard textbook. Within the SAE ap- tra (for details see [92]). proximation, however, our numerical scheme is tunable For ATI, the utilization of the 3D-TDSE [Eq. (8)] allow to treat any complex atom by choosing the adequate ef- us to calculate not only energy-resolved photoelectron fective (Hartree-Fock) potential VSAE(r), and finding the spectra, but also angular electron momentum distribu- ground state by the means of numerical diagonalization. tions of atoms driven by spatially inhomogeneous fields. Next, we will show how the inhomogeneity modifies As in the 1D case the nonhomogeneous character of the the equations which model the laser-electron coupling. laser electric field plays an important role on the ATI phe- Inserting Eq. (9) into Eq. (8) and considering that, nomenon. In addition, our 3D approach is able to model in a reliable way the ATI process both in the tunneling cos θY 0 = c Y 0 + c Y 0 (11) l l−1 l−1 l l+1 and multiphoton regimes. We show that for the former, and the spatial nonhomogeneous field causes significant modifications on the electron momentum distributions and 2 0 0 2 2 0 0 cos θYl = cl−2cl−1Yl−1 + (cl−1 + cl )Yl + clcl+1Yl+2, photoelectron spectra, while its effects in the later ap- (12) pear to be negligible. Indeed, through the tunneling ATI where process, one can obtain higher energy electrons as well as s a high degree of asymmetry in the momentum space map. (l + 1)2 In our study we consider NIR laser fields with intensities cl = , (13) (2l + 1)(2l + 3) in the mid- 1014 W cm−2 range. We use a linear approx- 7 imation for the plasmonic field, considered valid when different behavior in the spatially inhomogeneous field the electron excursion is small compared with the inho- when compared to the homogeneous field. In the case mogeneity region. Indeed, our 3D simulations confirm of inhomogeneous fields, the ionization and rescattering that plasmonic fields could drive electrons with energies times differ between neighboring cycles, despite the field in the near-keV regime (see e.g. [93]). being nearly monochromatic. Indeed, the contributions Similarly to the 1D case, the ATI spectrum is calcu- from one cycle may lead to a lower cutoff, while another lated starting from the time propagated electron wave may develop a higher cutoff. As was shown both by function, once the laser pulse has ceased. For comput- our quantum mechanical and classical models, our SFA ing the energy-resolved photoelectron spectra P (E) and model confirms that the ATI cutoff extends far beyond two-dimensional electron distributions H(kz, kr), where the semiclassical cutoff, as a function of inhomogeneity kz (kr) is the electron momentum component parallel strength. In addition, the angular momentum distribu- (perpendicular) to the polarization direction, we use the tions have very different features compared to the homo- window function approach developed in [89, 90]. geneous case. For the neighboring cycles, the electron Experimentally speaking, both the direct and rescat- momentum distributions do not share the same absolute tered electrons contribute to the energy-resolved photo- momentum, and as a consequence they do not have the electron spectra. It means that for tackling this problem same yield. both physical mechanisms should to be included in any theoretical model. In that sense, the 3D-TDSE, which can be considered as an exact approach to the ATI prob- C. Classical framework lem for atoms and molecules in the SAE approximation, appears to be the most suitable tool to predict the P (E) Important information such as the HHG cutoff and the in the whole range of electron energies. properties of the electron trajectories moving in the oscillatory laser electric field, can be obtained solving the classical one-dimensional Newton-Lorentz equation for B. Semiclassical approach an electron moving in a linearly polarized electric field. Specifically, we find the numerical solution of An independent approach to compute high-order har- x¨(t) = −∇xVl(x, t), (15) monic spectra for atoms in intense laser pulses is the Strong Field Approximation (SFA) or Lewenstein where Vl(x, t) is defined in Eq. (4) with the laser elec- model [17]. The main ingredient of this approach is the tric field linearly polarized in the x axis. For fixed values evaluation of the time-dependent dipole moment d(t). of ionization times ti, it is possible to obtain the clas- Within the SAE approximation, it can be calculated sical trajectories and to numerically calculate the times starting from the ionization and recombination transi- tr for which the electron recollides with the parent ion. tion matrices combined with the classical action of the In addition, once the ionization time ti is fixed, the full laser-ionized electron moving in the laser field. The SFA electron trajectory is completely determined (for more approximation has a direct interpretation in terms of the details about the classical model see [95]). so-called three-step or simple man’s model [15, 17]. The following conditions are commonly set (the result- Implicitly the Lewenstein model deals with spa- ing model is also known as the simple man’s model): i) tially homogeneous electric and vector potential fields, the electron starts with zero velocity at the origin at time i.e. fields that do not experience variations in the region t = ti, i.e., x(ti) = 0 andx ˙(ti) = 0; (ii) when the laser where the electron dynamics takes place. In order to con- electric field reverses its direction, the electron returns to sider spatial nonhomogeneous fields, the SFA approach its initial position, i.e., recombines with the parent ion, needs to be modified accordingly, i.e. the ionization and at a later time, t = tr, i.e. x(tr) = 0. ti and tr are known recombination transition matrices, joint with the classi- as ionization and recombination times, respectively. The cal action, now should take into account this new feature electron kinetic energy at tr can be obtained from the 2 of the laser electric and vector potential fields (for details usual formula Ek(tr) =x ˙(tr) /2, and, finding the value see [83, 94]). of tr (as a function of ti) that maximizes this energy, we As for the case of HHG driven by spatially inhomo- find that the HHG cutoff is given by ncω0 = 3.17Up + Ip, geneous fields, ATI can also be modeled by using the where nc is the harmonic order at the cutoff, ω0 is the SFA. In order to do so, it is necessary to modify the SFA laser frequency, Up is the ponderomotive energy and Ip ingredients, namely the classical action and the saddle is the ionization potential of the atom or molecule under point equations. The latter are more complex, but ap- consideration. It is worth mentioning that the HHG cut- pear to be solvable for the case of spatially linear inho- off will be extended when spatially inhomogeneous fields mogeneous fields (for details see [94]). Within SFA it is are employed. possible to investigate how the individual pairs of quan- From the simple-man’s model [15, 17] we can describe tum orbits contribute to the photoelectron spectra and the physical origin of the ATI process as follows: an the two-dimensional electron momentum distributions. atomic electron at a position x = 0, is released or We demonstrate that the quantum orbits have a very born at a given ionization time ti, with zero velocity, 8

FIG. 2: HHG spectra for a model atom with a ground-state energy, Ip = −0.67 a.u. obtained using the 1D-TDSE approach. The laser parameters are I = 2 × 1014 W·cm−2 and λ = 800 nm. We have used a trapezoidal shaped pulse with two optical cycles turn on and turn off, and a plateau with six optical cycles, 10 optical cycles in total, i.e. approximately 27 fs. The arrow indicates the cutoff predicted by the semiclassical model [17]. Panel (a): homogeneous case, (b): ε = 0.01 (100 a.u), (c): ε = 0.02 (50 a.u) and (d): ε = 0.05 (20 a.u). The numbers in brackets indicate an estimate of the inhomogeneity region (for more details see e.g [83, 85]) . In panels (e) and (f) is shown the dependence of the semiclassical trajectories on the ionization and recollision times for different values of ε and for the non confined case, panel (e) and the confined case, panel (f), respectively. Red squares: homogeneous case ε = 0; green circles: ε = 0.01; blue triangles: ε = 0.02 and blue triangles: ε = 0.05.

i.e.x ˙(ti) = 0. This electron now moves only under the III. HHG DRIVEN BY SPATIALLY influence of the oscillating laser electric field (the residual INHOMOGENEOUS FIELDS Coulomb interaction is neglected in this model) and will reach the detector either directly or through a rescatter- Field-enhanced high-order-harmonic generation ing process. By using the classical equation of motion, (HHG) using plasmonics fields, generated starting from it is possible to calculate the maximum energy of the engineered nanostructures or nanoparticles, requires electron for both direct and rescattered processes. no extra amplification stages due to the fact that, by For the direct ionization, the kinetic energy of an elec- exploiting surface plasmon resonances, the input driving tron released or born at time ti is electric field can be enhanced by more than 20 dB (corresponding to an increase in the intensity of 2-3 orders 2 of magnitude). As a consequence of this enhancement, [x ˙(ti) − x˙(tf )] Ed = , (16) the threshold laser intensity for HHG generation in 2 noble gases is largely exceeded and the pulse repetition rate remains unaltered. In addition, the high-harmonics where tf is the end time of the laser pulse. For the rescat- radiation generated from each nanosystem acts as a tering process, in which the electron returns to the core pointlike source, enabling a high collimation or focusing at a time tr and reverses its direction, the kinetic energy of this coherent radiation by means of (constructive) of the electron yields interference. This fact opens a wide range of possibilities to spatially arrange nanostructures to enhance or shape [x ˙(t ) +x ˙(t ) − 2x ˙(t )]2 the spectral and spatial properties of the harmonic E = i f r . (17) r 2 radiation in numerous ways [76, 78, 79]. Due to the nanometric size of the so-called plasmonic ’hot spots’, i.e. the spatial region where the electric field For homogeneous fields, Eqs. (16) and (17) become 2 2 reaches its highest intensity, one of the main theoreti- [A(ti)−A(tf )] [A(ti)+A(tf )−2A(tr )] Ed = 2 and Er = 2 , with cal assumptions, namely the spatial homogeneity of the A(t) being the laser vector potential A(t) = − R t E(t0)dt0. driven electric field, should be excluded. As a conse- For the case with ε = 0, it can be shown that the max- quence, both the analytical and numerical approaches to imum value for Ed is 2Up while for Er it is 10Up [28]. study laser-matter processes in atoms and molecules, in These two values appear as cutoffs in the energy-resolved particular HHG, need to be modified to treat adequately photoelectron spectrum. this different scenario and allow now for a spatial de- 9 pendence in the laser electric field. Several authors have B. Spatially (linear) nonhomogeneous fields: the addressed this problem recently [83–86, 88, 92–121]. As SFA approach we will show below, this new characteristic affects considerably the electron dynamics and this is reflected on In this subsection we summarize the work presented the observables, in the case of this subsection the HHG in [99]. In this contribution, we perform a detailed anal- spectra. ysis of high-order harmonic generation (HHG) in atoms within the strong field approximation (SFA) by considering spatially (linear) inhomogeneous monochromatic A. Spatially (linear) nonhomogeneous fields and laser fields. We investigate how the individual pairs of electron confinement quantum orbits contribute to the harmonic spectra. To this end we have modified both the classical action and In this sub-section we summarize the study carried out the saddle points equations by including explicitly the in [83] where it is demonstrated that both the inhomo- spatial dependence of the laser field. We show that in geneity of the local fields and the constraints in the elec- the case of a linear inhomogeneous field the electron tun- tron movement, play an important role in the HHG pro- nels with two different canonical momenta. One of these cess and lead to the generation of even harmonics and a momenta leads to a higher cutoff and the other one de- significant increase in the HHG cutoff, more pronounced velops a lower cutoff. Furthermore, we demonstrate that for longer wavelengths. In order to understand and char- the quantum orbits have a very different behavior in com- acterize these new HHG features we employ two of the parison to the conventional homogeneous field. A recent different approaches mentioned above: the numerical so- study supports our initial findings [122]. lution of the 1D-TDSE (see panels (a)-(d) in Fig. 2) and We also conclude that in the case of the inhomoge- the semiclassical approach known as Strong Field Ap- neous fields both odd and even harmonics are present in proximation (SFA). Both approaches predict comparable the HHG spectra. Within our extended SFA model, we results and describe satisfactorily the new features, but show that the HHG cutoff extends far beyond the stan- by employing the semiclassical arguments (see panels (e), dard semiclassical cutoff in spatially homogeneous fields. (f) in Fig. 2) behind the SFA and time-frequency analysis Our findings are in good agreement both with quantum- tools (Fig. 3), we are able to fully explain the reasons of mechanical and classical models. Furthermore, our ap- the cutoff extension. proach confirms the versatility of the SFA approach to tackle now the HHG driven by spatially (linear) inhomo-

&. &/ +" +" '" +" geneous ﬁelds. !* )" !* '" )" %" )" %"

#" #" '" $*" %$" %(" '" $*" %$" %(" &"" !+ !+

,.746530%67127 %" ,.746530%67127 %" C. Real nonhomogeneous ﬁelds

#" !#" #" !#" " #"" %"" '"" )"" +"" ##"" " #"" %"" '"" )"" +"" ##"" -342%&.!8! -342%&.!8! In this sub-section we present numerical simulations of HHG in an argon model atom produced by the ﬁelds gen-

&0 +" &1 +" +" +" erated when a gold bow-tie nanostructure is illuminated )" !* )" !* '" '" by a short laser pulse of long wavelength λ = 1800 nm )" )" %" %" #" #" (see [84] for more details). The functional form of these '" $*" %$" %(" &"" '" $*" %$" %(" &"" !+ !+ fields is extracted from finite element simulations using ,.746530%67127 %" ,.746530%67127 %" both the complete geometry of the metal nanostructure #" !#" #" !#" " #"" %"" '"" )"" +"" ##"" " #"" %"" '"" )"" +"" ##"" and laser pulse characteristics (see Fig. 4(a)). We use -342%&.!8! -342%&.!8! the numerical solution of the TDSE in reduced dimen- FIG. 3: Panels (a)-(d): Gabor analysis for the sions to predict the HHG spectra. A clear extension in corresponding HHG spectra of panels (a)-(d) of Fig. 2. the harmonic cutoff position is observed. This character- The zoomed regions in all panels show a time interval istic could lead to the production of XUV coherent laser during the laser pulse for which the complete electron sources and open the avenue to the generation of shorter trajectory, from birth time to recollision time, falls attosecond pulses. It is shown in Fig. 4(c) that this new within the pulse plateau. In panels (a)-(d) the color feature is a consequence of the combination of a spatial scale is logarithmic. nonhomogeneous electric field, which modifies substantially the electron trajectories, and the confinement of the electron dynamics. Furthermore, our numerical results are supported by time-analysis and classical simulations. A more pronounced increase in the harmonic cutoff, in addition to an appreciable growth in conversion efficiency, could be attained by optimizing the nanostructure geometry and materials. These degrees of freedom could pave the way to tailor the harmonic spectra ac- 10

(c) as produced after the interaction of the laser pulse with nanoplasmonic antennas [76, 83, 85, 86], metallic nanowaveguides [78], metal [62, 124] and dielectric nanoparticles [125] or metal nanotips [63, 64, 126–129]. The coupling between the atom and the laser pulse, linearly polarized along the z axis, is modified in order to treat the spatially nonhomogeneous fields and can be written it as: Vl(z, t, τ) = E˜(z, t, τ) z with E˜(z, t, τ) = E(t, τ)(1 + εz) and E(t, τ) = E1(t) + E2(t, τ) the tempo- FIG. 4: (a) Schematic representation of the geometry of ral synthesized laser field with τ the time delay between the considered nanostructure. A gold bow-tie antenna the two pulses (see e.g. [123] for more details). As in the resides on glass substrate (refractive index n = 1.52) 1D case the parameter ε defines the strength of the non- with superstate medium of air (n = 1). The homogeneity and the dipole approximation is preserved characteristic dimensions of the system and the because ε 1. coordinate system used in the 1D-TDSE simulations are shown. (b) SEM image of a real gold bow-tie antenna. (c) High-order harmonic generation (HHG) spectra for (a) a model of argon atoms (Ip = −0.58 a.u.), driven by a laser pulse with wavelength λ = 1800 nm and intensity I = 1.25 × 1014 W·cm−2 at the center of the gap x = 0. We have used a trapezoidal shaped pulse with three

optical cycles turn on and turn oﬀ, and a plateau with (eV) energy four optical cycles (about 60 fs). The gold bow-tie nanostructure has a gap g = 15 nm (283 a.u.). The black line indicates the homogeneous case while the red line indicates the nonhomogeneous case. The arrow optical cycles indicates the cutoﬀ predicted by the semiclassical model !4 for the homogeneous case [17]. The top left inset shows !6 (b) 3.17Up

the functional form of the electric ﬁeld E(x, t), where ) 2 (a.u.) (a.u.) !8 )|

2 2 12.5Up the solid lines are the raw data obtained from the ﬁnite ! element simulations and the dashed line is a nonlinear !10 log(|a( ﬁtting. The top right inset shows the intensity !12

enhancement in the gap region of the gold bow-tie la(w)l log !14

nanostructure. 0 200 400 600 800 1000 1200 Photonenergy energy (eV) (eV) . cording to specific requirements. FIG. 5: (a) Time-frequency analysis obtained from the 3D-TDSE harmonic spectrum for a He atom driven by D. Temporal and spatial synthesized fields the spatially and temporally synthesized pulse described in the text with ε = 0.002. The plasmonic enhanced intensity I = 1.4 × 1015 W cm−2. Superimposed (in In this sub-section we present a brief summary of the brown) are the classical rescattering energies; (b) results published in [92]. In short, numerical simulations 3D-TDSE harmonic spectrum for the same parameters of HHG in He atoms using a temporal and spatial synthe- used in (a). sized laser field are considered using the full 3D-TDSE. This particular field provides a new route for the generation of photons at energies beyond the carbon K-edge The linear functional form for the spatial non- using laser pulses at 800 nm, which can be obtained from homogeneity described above could be obtained engineer- conventional Ti:Sapphire laser sources. The temporal ing adequately the geometry of plasmonic nanostructures synthesis is performed using two few-cycle laser pulses and by adjusting the laser parameters in such a way that delayed in time [123]. On the other hand, the spatial the laser-ionized electron feels only a linear spatial vari- synthesis is obtained by using a spatial nonhomogeneous ation of the laser electric field when in the continuum laser field [83, 85, 86] produced when a laser beam is fo- (see e.g. [84] and references therein). The harmonic spec- cused in the vicinity of a metal nanostructure or nanopar- trum then obtained in He for ε = 0.002 is presented in ticle. Fig. 5(b). We can observe a considerable cut-off exten- Focusing on the spatial synthesis, the nonhomoge- sion up to 12.5Up which is much larger when compared neous spatial distribution of the laser electric field can with the double pulse configuration employed alone (it be obtained experimentally by using the resulting field leads only to a maximum of 4.5Up [123]). This large ex- 11 tension of the cutoff is therefore a signature of the com- of the atom is obtained by Fourier transforming the ac- bined effect of the double pulse and the spatial nonho- celeration a(t) of the electronic wavepacket. mogeneous character of the laser electric field. For this Figure 6, panels (a), (b) and (c) show the harmonic 15 particular value of the laser peak intensity (1.4 × 10 W spectra for model xenon atoms generated by a laser pulse cm−2) the highest photon energy is greater than 1 keV. 13 −2 with I = 2×10 W cm , λ = 720 nm and a τL = 13 fs, Note that the quoted intensity is actually the plasmonic i.e. np = 5 (which corresponds to an intensity envelope enhanced intensity, not the input laser intensity. The lat- of ≈ 4.7 fs FWHM) [124]. In the case of a spatial ho- ter could be several orders of magnitude smaller, accord- mogeneous field, no harmonics beyond the 9th order are ing to the plasmonic enhancement factor (see e.g. [76, 78]) observed. The spatial decay parameter χ accounts for and will allow the nanoplasmonic target to survive to the the spatial nonhomogeneity induced by the nanoparticle interaction. In order to confirm the underlying physics and it varies together with its size and the kind of metal highlighted by the classical trajectories analysis, we have employed. Varying the value of χ is therefore equivalent retrieved the time-frequency distribution of the calcu- to choosing the type of nanoparticle used, which allows lated dipole (from the 3D-TDSE) corresponding to the to overcome the semiclassically predicted cutoff limit and case of the spectra presented in Fig. 5(b) using a wavelet reach higher harmonic orders. For example, with χ = 40 analysis. The result is presented in Fig. 5(a) where we and χ = 50 harmonics in the mid 20s (panel c) and well have superimposed the calculated classical recombination th above the 9 (a clear cutoff at nc ≈ 15 is achieved) (panel energies (in brown) to show the excellent agreement be- b), respectively, are obtained. A modification in the har- tween the two theoretical approaches. The consistency of monic periodicity, related to the breaking of symmetry the classical calculations with the full quantum approach imposed by the induced nonhomogeneity, is also clearly is clear and confirms the mechanism of the generation of noticeable. this 12.5Up cut-off extension. In addition, the HHG spectra exhibit a clean continuum as a result of the trajectory selection on the recombination time, which itself is a consequence of employing a combination of temporally and spatially synthesized laser field.

E. Plasmonic near-ﬁelds

This sub-section includes an overview of the results reported in [88]. In this contribution it is shown how the HHG spectra from model Xe atoms are modified by using a plasmonic near enhanced field generated when a metal nanoparticle is illuminated by a short laser pulse. A setup combining a noble gas as a driven media and metal nanoparticles was also proposed recently in [108, 130]. FIG. 6: HHG spectra for model Xe atoms, laser wavelength λ = 720 nm and intensity I = 2 × 1013 For our near-field we use the function given by [124] −2 2 to define the spatial nonhomogeneous laser electric field W·cm . We use a sin pulse envelope with n = 5. E(x, t), i.e. Panel (a) represents the homogeneous case, panel (b) χ = 50 and panel (c) χ = 40. The arrow in panel (a) E(x, t) = E0 f(t) exp(−x/χ) sin(ω0t + φ), (18) indicates the cutoff predicted by the semiclassical approach [17]. Panels (d), (e), (f) show the where E0, ω0, f(t) and φ are the peak amplitude, the laser field frequency, the field envelope and the CEP, re- corresponding total energy of the electron (expressed in spectively. The functional form of the resulting laser elec- harmonic order) driven by the laser field calculated tric field is extracted from attosecond streaking experi- from the one-dimensional Newton-Lorentz equation and ments and incorporated both in our quantum and clas- plotted as a function of the ti (green (light gray) circles) sical approaches. In this specific case the spatial depen- or the tr (red (dark gray) circles). dence of the plasmonic near-field is given by exp(−x/χ) and it is a function of both the size and the material of the Now, by the semiclassical simple man’s (SM) spherical nanoparticle. E(x, t) is valid for x outside of the model [15, 17] we will study the harmonic cut-off exten- metal nanoparticle, i.e. x ≥ R , where R is its radius. It 0 0 sion. This new effect may be caused by a combination of is important to note that the electron motion takes place several factors (for details see [83, 84]). As is well known, in the region x ≥ R0 with (x + R0) 0. We consider the cutoff law is nc = (3.17Up + Ip)/ω0, where nc is the the laser field having a sin2 envelope: f(t) = sin2 ω0t , 2np harmonic order at the cutoff and Up the ponderomotive where np is the total number of optical cycles, i.e. the to- energy. We solve numerically Eq. (15) for an electron tal pulse duration is τL = 2πnp/ω0. The harmonic yield moving in an electric field with the same parameters used 12 in the TDSE-1D calculations, i.e. quently electrons with high kinetic energy are needed in order to describe it [28, 30, 132]. x(t) x¨(t) = −∇ V (x, t) = −E(x, t) 1 − , (19) Nowadays, experiments have demonstrated that ATI x l χ photoelectron spectra could be extended further by using plasmon-field enhancement [62, 76]. The strong con- and consider the SM model initial conditions: the elec- finement of the plasmonics spots and the distortion of tron starts at position zero at t = ti (the ionization time) the electric field by the surface plasmons induces a spa- with zero velocity, i.e. x(ti) = 0 andx ˙(ti) = 0. When tial inhomogeneity in the driving laser field, just be- the electric field reverses, the electron returns to its ini- fore the interaction with the corresponding target gas. tial position (i.e. the electron recollides or recombines A related process employing solid state targets instead with the parent ion) at a later time t = tr (the recombi- of atoms and molecules in gas phase is the so called nation time), i.e. x(tr) = 0. The electron kinetic energy above-threshold photoemission (ATP). This laser driven 2 x˙ (tr ) phenomenon has received special attention recently due at the tr is calculated as usual from: Ek(tr) = 2 and finding the tr (as a function of ti) that maximizes Ek, nc to its novelty and the new physics involved. In ATP is also maximized. electrons are emitted directly from metallic surfaces or Panels (d), (e) and (f) of Fig. 6 represent the behaviour metal nanotips and they present distinct characteris- of the harmonic order upon the ti and tr, calculated from tics, namely higher energies, far beyond the usual cut- n = (Ek(ti,r) + Ip)/ω0 as for the cases (a), (b) and (c) of off for noble gases and consequently the possibility to Fig. 6, respectively. Panels (e) and (f) show how the non- reach similar electron energies with smaller laser inten- homogeneous character of the laser field strongly modi- sities (see e.g. [63, 64, 126, 127, 133]). Furthermore, the fies the electron trajectories towards an extension of the photoelectrons emitted from these nanosources are sensi- nc. This is clearly present at nc ∼ 18ω0 (28 eV) and tive to the CEP and consequently it plays an important nc ∼ 27ω0 (42 eV) for χ = 50 and χ = 40, respectively. role in the angle and energy resolved photoelectron spec- These last two cutoff extensions are consistent with the tra [62, 63, 134]. quantum predictions presented in panels (b) and (c) of Fig. 6. Classical and quantum approaches predict cutoff ex- A. ATI driven by spatially linear inhomogeneous tensions that could lead to the production of XUV co- fields: the 1D-case herent laser sources and open a direct route to the generation of attosecond pulses. This effect is caused by the For our 1D quantum simulations we employ as a driv- induced laser field spatial nonhomogeneity, which modi- ing field a four-cycle (total duration 10 fs) sin-squared fies substantially the electron trajectories. A more pro- laser pulse with an intensity I = 3 × 1014 W cm−2 and nounced increment in the harmonic cutoff, in addition to wavelength λ = 800 nm. We chose a linear inhomoge- an appreciable growth in the conversion efficiency, could neous field and three different values for the parameter be reached by varying both the radius and the metal that characterizes the inhomogeneity strength, namely material of the spherical nanoparticles. These new de- ε = 0 (homogeneous case), ε = 0.003 and ε = 0.005. grees of freedom could pave the way to extend the har- Figure 7(a) shows the cases with φ = 0 (a sin-like laser monic plateau reaching the XUV regime with modest in- pulse) meanwhile in Fig. 7(b) φ = π/2 (a cos-like laser put laser intensities. pulse), respectively. In both panels green represents the homogeneous case, i.e. ε = 0, magenta is for ε = 0.003 and yellow is for ε = 0.005, respectively. IV. ATI DRIVEN BY SPATIALLY For the homogeneous case, the spectra exhibits the INHOMOGENEOUS FIELDS usual distinct behavior, namely the 2Up cutoff (≈ 36 eV for our case) and the 10Up cutoff (≈ 180 eV), where 2 2 As was mentioned at the outset, ATI represents an- Up = E0 /4ω0 is the ponderomotive potential. The for- other fundamental strong field phenomenon. Investiga- mer cutoff corresponds to those electrons that, once ion- tions carried out on ATI, generated by few-cycle driving ized, never return to the atomic core, while the latter laser pulses, have attracted much interest due to the sen- one corresponds to the electrons that, once ionized, re- sitivity of the energy and angle-resolved photoelectron turn to the core and elastically rescatter. It is well estab- spectra to the absolute value of the CEP [28, 131]. This lished using classical arguments that the maximum ki- feature makes the ATI phenomenon a conceivable tool for netic energies of the direct and the rescattered electrons d r laser pulse characterization. In order to characterize the are Emax = 2Up and Emax = 10Up, respectively. In a CEP of a few-cycle laser pulse, the so-called backward- quantum mechanical approach, however, it is possible to forward asymmetry of the ATI spectrum is measured and find electrons with energies beyond the 10Up, although from the information collected the absolute CEP value their yield drops several orders of magnitude [28]. The can be obtained [131, 132]. Furthermore, nothing but the TDSE, which can be considered as an exact approach to high energy region of the photoelectron spectrum appears the problem, is able to predict the P (E) for the whole to be strongly sensitive to the absolute CEP and conse- range of electron energies. In addition, the most energetic 13

0 0 of motion for the electron in the laser ﬁeld can be written 2Up =0 2Up = /2 ! (a) ! ! (b) as (see Eq. (15)): !5 !5 x¨(t) = −∇xVl(x, t) 10Up 10Up = E(x, t) + [∇xE(x, t)] x !10 !10 = E(t)(1 + 2εx(t)), (20) where we have collected the time dependent part of the !15

log electron yield (arb. units) !15 0 500 log electron yield (arb. units) 0 500 electric field in E(t), i.e. E(t) = E0f(t) sin(ω0t + φ) and Photoelectron energy (eV) Photoelectron energy (eV) particularized to the case h(x) = x. In the limit where ε = 0 in Eq. (20), we recover the spatial homogeneous FIG. 7: 1D-TDSE energy-resolved photoelectron case. Using the classical formalism described in Sec- spectra for a model atom with Ip = −0.5 a.u. and for 14 −2 tion II C. we find the maximum energy for both the direct the laser parameters, I = 3 × 10 W cm , λ = 800 nm and rescattered electrons. As can be seen, the electron and a sin-squared shaped pulse with a total duration of energy cutoffs now exceed the ones obtained for conven- 4 cycles (10 fs). In green for ε = 0 (homogeneous case), tional fields (see panels (a) and (b), in green, in Fig. 7 in magenta for ε = 0.003 and in yellow for ε = 0.005. and the respective arrows). Panel (a) represent the case for φ = 0 (sin-like pulse) and panel (b) represents the case for φ = π/2 (cos-like 200 pulse). The arrows indicate the 2Up and 10Up cutoffs (a) !=0 (d) !=!/2 predicted by the classical model [28] 150 "=0 "=0

100 electrons, i.e. those with Ek 2Up, are used to charac- 50 terize the CEP of few-cycle pulses. As a result, a correct description of the rescattering mechanism is needed. 2500 (b) =0 (e) !=!/2 For the spatial inhomogeneous case, the cutoﬀ posi- 200 ! tions of both the direct and the rescattered electrons "=0.003 "=0.003 are extended towards larger energies. For the rescattered 150 electrons, this extension is very prominent. In fact, for 100

ε = 0.003 and ε = 0.005, it reaches ≈ 260 eV and ≈ 420 50 eV, respectively (see Fig. 7(a)). Furthermore, it appears electron energy (eV) that the high energy region of P (E), for instance, the 3500 (c) !=0 (f) !=!/2 region between 200 − 400 eV for ε = 0.005 (Fig. 7 in 280 =0.005 yellow), is strongly sensitive to the CEP. This feature in- "=0.005 " 210 dicates that the high energy region of the photoelectron spectra could resemble a new and better CEP charac- 140 terization tool. It should be, however, complemented by 70 other well known and established CEP characterization 0 tools, as, for instance, the forward-backward asymmetry 0 1 2 3 40 1 2 3 4 (see [28]). Furthermore, the utilization of nonhomoge- optical cycles neous fields would open the avenue for the production of high energy electrons, reaching the keV regime, if a FIG. 8: Numerical solutions of Eq. (20) plotted in terms reliable control of the spatial and temporal shape of the of the direct (blue) and rescattered (orange) electron laser electric field is attained. kinetic energy. The laser parameters are the same as in We now concentrate our efforts on explaining the Fig. 7. Panels (a), (b) and (c) correspond to the case of extension of the energy-resolved photoelectron spec- sin-like pulses (φ = 0) and for ε = 0 (homogeneous tra using classical arguments. From the simple-man’s case), ε = 0.003 and ε = 0.005, respectively. Panels (d), model [15, 17] we can describe the physical origin of the (e) and (f) correspond to the case of cos-like pulses ATI process as follows: an atomic electron at a posi- (φ = π/2) and for ε = 0 (homogeneous case), ε = 0.003 tion x = 0, is released or born at a given time, that we and ε = 0.005, respectively. call ionization time ti, with zero velocity, i.e.x ˙(ti) = 0. This electron now moves only under the influence of the In Fig. 8, we present the numerical solutions of oscillating laser electric field (the residual Coulomb in- Eq. (20), which is plotted in terms of the kinetic energy teraction is neglected in this model) and will reach the of the direct and rescattered electrons. We employ the detector either directly or through a rescattering process. same laser parameters as in Fig. 7. Panels (a), (b) and (c) By using the classical equation of motion, it is possible correspond to the case of φ = 0 (sin-like pulses) and for to calculate the maximum energy of the electron for both ε = 0 (homogeneous case), ε = 0.003 and ε = 0.005, re- direct and rescattered processes. The Newton equation spectively. Meanwhile, panels (d), (e) and (f) correspond 14 to the case of φ = π/2 (cos-like pulses) and for ε = 0 (ho- By a simple inspection of Fig. 9 strong modifications pro- mogeneous case), ε = 0.003 and ε = 0.005, respectively. duced by the spatial inhomogeneities in both the angular From the panels (b), (c), (e) and (f) we can observe the and low-energy structures can be appreciated (see [93] strong modifications that the nonhomogeneous character for more details). of the laser electric field produces in the electron kinetic

2 energy. These are related to the changes in the electron (a) trajectories (for details see e.g. [83, 84, 86]). In short, −2 the electron trajectories are modified in such a way that −4 1 (a.u.) r now the electron ionizes at an earlier time and recom- k −6 bines later, and in this way it spends more time in the 0 −8 continuum acquiring energy from the laser electric field. −2 −1 0 1 2 k (a.u.) Consequently, higher values of the kinetic energy are at- z 2 (b) tained. A similar behavior with the photoelectrons was −2 observed recently in ATP using metal nanotips. Accord- −4 1 ing to the model presented in [64] the localized fields (a.u.) r k modify the electron motion in such a way to allow sub- −6 cycle dynamics. In our studies, however, we consider 0 −8 −2 −1 0 1 2 both direct and rescattered electrons (in [64] only direct k (a.u.) z electrons are modeled) and the characterization of the dynamics of the photoelectrons is more complex. Never- FIG. 10: (Color online) Two-dimensional electron theless, the higher kinetic energy of the rescattered elec- momentum distributions (logarithmic scale) in trons is a clear consequence of the strong modifications cylindrical coordinates (kz, kr) using the exact of the laser electric field in the region where the electron 3D-TDSE calculation for an hydrogen atom. The laser 13 dynamics takes place, as in the above mentioned case of parameters are E0 = 0.05 a.u. (I = 8.775 × 10 W −2 ATP. cm ), ω0 = 0.25 a.u. (λ = 182.5 nm) and φ = π/2. We employ a laser pulse with 6 total cycles. Panel (a) corresponds to the homogeneous case (ε = 0) and panel B. ATI driven by spatially linear inhomogeneous (b) is for ε = 0.005. fields: the 3D-case However in the case of low intensity regime (i.e. multiphoton regime, γ 1) the scenario changes radically.

2 −1 2 −1 In order to study this regime we use a laser electric ﬁeld (a) (b) − − 13 2 2 with E0 = 0.05 a.u. of peak amplitude (I = 8.775 × 10 −3 −3 1 1 −2 (a.u.) (a.u.) r −4 r −4 W cm ), ω = 0.25 a.u. (λ = 182.5 nm) and 6 com- k k 0 −5 −5 plete optical cycles. The resulting Keldysh parameter 0 −6 0 −6 −2 −1 0 1 2 −2 −1 0 1 2 k (a.u.) k (a.u.) z z γ = 5 indicates the predominance of a multiphoton pro-

2 −1 2 −1 cess [135]. In Fig. 10 we show the two-dimensional elec- (c) (d) −2 −2 tron distributions for the two cases discussed above. For −3 −3 1 1 (a.u.) (a.u.) r −4 r −4

k k the homogeneous case our calculation is identical to the −5 −5 one presented in [135]. We also notice the two panels 0 −6 0 −6 −2 −1 0 1 2 −2 −1 0 1 2 k (a.u.) k (a.u.) z z present indistinguishable shape and magnitude. Hence the differences introduced by the spatial inhomogeneity FIG. 9: (Color online) Two-dimensional electron are practically imperceptible in the multiphoton ioniza- momentum distributions (logarithmic scale) in tion regime. cylindrical coordinates (kz, kr) using the exact 3D-TDSE calculation for an hydrogen atom. The laser 14 −2 parameters are I = 5.0544 × 10 W cm (E0 = 0.12 C. Plasmonic near-fields a.u.) and λ = 800 nm. We have used a sin-squared shaped pulse with a total duration of four optical cycles In this section we put forward the plausibility to per- (10 fs) with φ = π/2. (a) ε = 0 (homogeneous case), (b) form ATI experiments by combining plasmonic enhanced ε = 0.002, (c) ε = 0.003 and (d) ε = 0.005. near-fields and noble gases. The proposed experiment would take advantage of the plasmonic enhanced near- In the following, we calculate two-dimensional elec- fields (also known as evanescent fields), which present a tron momentum distributions for a laser field intensity of strong spatial nonhomogeneous character and the flex- 14 −2 I = 5.0544 × 10 W cm (E0 = 0.12 a.u). The results ibility to use any atom or molecule in gas phase. A are depicted in Fig. 9 for φ = π/2. Here, panels (a), (b), similar scheme was previously presented, but now we (c) and (d) represent the cases with ε = 0 (homogeneous are interested in generating highly energetic electrons, case), ε = 0.002, ε = 0.003 and ε = 0.005, respectively. instead of coherent electromagnetic radiation. We em- 15 ploy 1D-TDSE by including the actual functional form of other laser parameters fixed. From this plot we observe metal nanoparticles plasmonic near-fields obtained from that the nonhomogeneous character of the laser enhanced attosecond streaking measurements. We have chosen this electric field introduces a highly nonlinear behavior. For particular nanostructure since its actual enhanced-field this intensity with χ = 29 it is possible to obtain very is known experimentally, while for the other nanostruc- energetic electrons reaching values of several hundreds of tures, like bow-ties [76], the actual plasmonic field is un- eV. This is a good indication that the nonlinear behavior known. For most of the plasmonic nanostructures the en- of the combined system of the metallic nanoparticles and hanced field is theoretically calculated using the finite el- noble gas atoms could pave the way to generate keV elec- ement simulation, which is based on an ideal system that trons with tabletop laser sources. All the above quantum may deviate significantly from actual experimental condi- mechanical predictions can be directly confirmed by us- tions. For instance, [76] states an intensity enhancement ing classical simulations in the same way as for the case of 4 orders of magnitude (calculated theoretically) but HHG. the maximum harmonic measured was the 17th, which corresponds to an intensity enhancement of only 2 or- !2 0 (a) !=0 (b) !=0 ders of magnitude (for more details see [84, 101]). On !=40 !=29 !=30 !2 the other hand, our numerical tools allow a treatment !4 !=29 of a very general set of spatial nonhomogeneous fields !4 such as those present in the vicinity of metal nanostruc- !6 tures [76], dielectric nanoparticles [62], or metal nan- !6 !8 otips [64]. The kinetic energy for the electrons both !8 direct and rescattered can be classically calculated and !10 log electron yield (arb. units) !10 25 log electron yield (arb. units) compared to quantum mechanical predictions (for more 0 50 1 2 10 100 4 1000 details see e.g [104]). Photoelectron energy (eV) Photoelectronlog Phot. energy energy (eV) (eV) We have employed the same parameters as the ones used in Section III E, but now our aim is to compute FIG. 11: Energy resolved photoelectron spectra for Xe the energy resolved photoelectron spectra. In Fig. 11 we atoms driven by an electric enhanced near-field. In present the photoelectron spectra calculated using 1D- panel (a) the laser intensity after interacting with the 13 −2 TDSE for Xe atoms and for two different laser inten- metal nanoparticles is I = 2 × 10 W cm . We sities, namely I = 2 × 1013 W cm−2 (Fig. 11(a)) and employ φ = π/2 (cos-like pulses) and the laser I = 5 × 1013 W cm−2 (Fig. 11(b)). In Fig. 11(a) each wavelength and number of cycles remain unchanged curve presents different values of χ: homogeneous case with respect to the input pulse, i.e. λ = 720 nm and (χ → ∞), χ = 40, χ = 35 and χ = 29. For the homoge- np = 5 (13 fs in total). Panel (b) shows the output laser 13 −2 neous case there is a visible cutoff at ≈ 10.5 eV confirm- intensity of I = 5 × 10 W cm (everything else is the same as in panel (a)). The arrow indicates two ing the well known ATI cutoff at 10Up, which corresponds to those electrons that once ionized return to the core and conventional classical limits: 2Up (in red) at 5.24 eV and 10U (in blue) at 26.2 eV, respectively. elastically rescatter. Here, Up is the ponderomotive po- p 2 2 tential given by Up = Ep /4ω0. On the other hand, for this particular intensity, the cutoff at 2Up (≈ 2.1 eV) de- Here we propose generation of high energy photoelec- veloped by the direct ionized electrons is not visible in trons using near-enhanced fields by combining metallic the spectrum. nanoparticles and noble gas atoms. Near-enhanced fields For the spatial nonhomogeneous cases the cutoff of the present a strong spatial dependence at a nanometer scale rescattered electron is far beyond the classical limit 10Up, and this behavior introduces substantial changes in the depending on the χ parameter chosen. As it is depicted in laser-matter processes. We have modified the 1D-TDSE Fig. 11(a) the cutoff is extended as we decrease the value to model the ATI phenomenon in noble gases driven by of χ. For χ = 40 the cutoff is at around 14 eV, while for the enhanced near-fields of such nanostructure. We pre- χ = 29 it is around 30 eV. The low energy region of the dict a substantial extension in the cutoff position of the photoelectron spectra is sensitive to the atomic potential energy-resolved photoelectron spectra, far beyond the of the target and one needs to calculate TDSE in full conventional 10Up classical limit. These new features dimensionality in order to model this region adequately. are well reproduced by classical simulations. Our pre- In this paper we are interested in the high energy region dictions would pave the way to the production of high of the photoelectron spectra, which is very convenient energy photoelectrons reaching the keV regime by using because it is not greatly affected by the considered atom. a combination of metal nanoparticles and noble gases. In Thus by employing 1D-TDSE the conclusions that can this kind of system each metal nanoparticle configures a be taken from these highly energetic electrons are very laser nanosource with particular characteristics that al- reliable. low not only the amplification of the input laser field, Figure 11(b) shows the photoelectron spectra for the but also the modification of the laser-matter phenomena homogeneous case and for χ = 29 using a larger laser due to the strong spatial dependence of the generated field intensity of I = 5 × 1013 W cm−2, while keeping all coherent electromagnetic radiation. 16

D. Emergence of a higher energy structure (HES) processes driven by them. in ATI driven by spatially inhomogeneous laser fields Theoretically speaking, in the recent years there has been a thoughtful and continuous activity in atto- Our final example deals with a recent study about the nanophysics. Indeed, all of the theoretical tools devel- appearance of a higher energy structure (HES) in the oped to tackle strong field processes driven by spatially energy-resolved ATI photoelectron spectra when the ac- homogeneous fields have beed generalized and adapted tive media is driven by a spatially inhomogeneous laser to this new arena. Several open problems, however, still field [136]. As was discussed throughout this contribution remain. For instance, the behaviour of complex systems, the theoretical approaches had not considered any spatial e.g. multielectronic atoms and molecules, under the in- dependence in the field (forces) experienced by the laser- fluence of spatial inhomogeneous fields is an unexplored ionized electron. On the other hand, the small spatial area – only few attempts to tackle this problem has been inhomogeneity introduced by the long-range Coulomb recently reported [97, 137, 138]. In addition, and just potential has been recently linked to a number of im- to name another example, it was recently demonstrated portant features in the photoelectron spectrum, such as that Rydberg atoms could be a plausible alternative as a Coulomb asymmetry, Coulomb focusing, and a distinct driven media [139]. set of low energy structures in the angle-resolved pho- Diverse paths could be explored in the future. The ma- toelectron spectra. We demonstrated that using a mid- nipulation and control of the plasmonic-enhanced fields infrared laser source, with a time-varying spatial depen- appears as one them. From an experimental perspec- dence in the laser electric field, such as that produced tive this presents a tremendous challenge, considering the in the vicinity of a nanostructure, creates a prominent nanometric dimensions of the systems, although several higher energy peak. This HES originates from direct elec- experiments are planned in this direction, for instance trons ionized near the peak of a single half-cycle of the combining metal nanotips and molecules in a gas phase. laser pulse. This feature is indeed confirmed both using The possibility to tailor the electron trajectories at their quantum mechanical, TDSE-based, and classical, classi- natural scale is another path to be considered. By em- cal trajectory monte carlo (CTMC), approaches. Inter- ploying quantum control tools it would be possible, in estingly, the HES is well separated from all other ion- principle theoretically, to drive the electron following a ization events, with its location and energy width are certain desired ’target’,e.g. a one which results with the strongly dependent on the properties of the spatial inho- largest possible velocity, now with a time and spatial de- mogeneous field. As a consequence, the HES can be em- pendent driving field. The spatial shape of this field could ployed as a sensitive tool for near-field characterization in be, subsequently, obtained by engineering a nanostruc- a regime where the electron’s quiver amplitude is on the ture. order to the field decay length. Additionally, the large ac- The quest for HHG from plasmonic nano-structures, cumulation of electrons with tuneable energy suggests a joint with an explosive amount of theoretical work, begun promising method for creating a localized source of elec- with the controversial report of a Korean group on HHG tron pulses of sub-femtosecond duration using tabletop from bow-tie metal nano structures [76]. Let us men- laser technology. tion at the end a very recent results of the same group, which clearly seems to be well justified and, as such, opens new perspectives and ways toward efficient HHG V. CONCLUSIONS, OUTLOOK AND in nano-structures. In this recent article the authors PERSPECTIVES demonstrate plasmonic HHG experimentally by devising a metal-sapphire nanostructure that provides a solid tip In this contribution we have extensively reviewed the as the HHG emitter instead of gaseous atoms. The fab- theoretical tools to tackle strong field phenomena driven ricated solid tips are made of monocrystalline sapphire by plasmonic-enhanced fields and discussed a set of rele- surrounded by a gold thin-film layer, and intended to pro- vant results. duce coherent extreme ultraviolet (XUV) harmonics by Nowadays, for the first time in the history of Atomic the inter- and intra-band oscillations of electrons driven Molecular and Optical (AMO) physics we have at our by the incident laser. The metal-sapphire nanostructure disposal laser sources, which, combined with nanostruc- enhances the incident laser field by means of surface plas- tures, generate fields that exhibit spatial variation at a mon polaritons (SPPs), triggering HHG directly from nanometric scale. This is the native scale of the electron moderate femtosecond pulses of 0.1 TW cm−2 intensi- dynamics in atoms, molecules and bulk matter. Conse- ties. Measured XUV spectra show odd-order harmonics quently, markedly and profound changes occur in systems up to 60 nm wavelengths without the plasma atomic lines interacting with such spatially inhomogeneous fields. Us- typically seen when using gaseous atoms as the HHG ing well-known numerical techniques, based on solutions emitter. This experimental outcome confirms that the of Maxwell equations, one is able to model both the time plasmonic HHG approach is a promising way to realize and the spatial properties of these laser induced plas- coherent XUV sources for nano-scale near-field applica- monic fields. This in the first important step for the sub- tions in spectroscopy, microscopy, lithography, and at- sequent theoretical modelling of the strong-field physical tosecond physics [140]. The era of the atto-nanophysics 17 has just started. Fund. M. L. acknowledges the Spanish Ministry MINECO (National Plan 15 Grant: FISICATEAMO No. FIS2016-79508-P, SEVERO OCHOA No. SEV- VI. ACKNOWLEDGMENTS 2015-0522, FPI), European Social Fund, FundacióCellex, Generalitat de Catalunya (AGAUR Grant No. 2017 This work was supported by the project ELI–Extreme SGR 1341 and CERCA/Program), ERC AdG OSYRIS Light Infrastructure–phase 2 (CZ.02.1.01/0.0/0.0/15 and NOQIA, and the National Science Centre, Poland- 008/0000162) from European Regional Development Symfonia Grant No. 2016/20/W/ST4/00314.

[1] A. Scrinzi, M. Y. Ivanov, R. Kienberger, and D. M. [16] K. J. Schafer, B.Yang, L. F. DiMauro, and K. C. Kulan- Villeneuve, “Attosecond physics,” J. Phys. B 39, R1– der, “Above threshold ionization beyond the high har- R37 (2006). monic cutoff,” Phys. Rev. Lett. 70, 1599 (1993). [2] F. Krausz and M. Ivanov, “Attosecond physics,” Rev. [17] M. Lewenstein, P. Balcou, M. Y. Ivanov, A. L’Huillier, Mod. Phys. 81, 163–234 (2009). and P. B. Corkum, “Theory of high-harmonic genera- [3] M. Hentschel, R. Kienberger, C. Spielmann, G. A. Rei- tion by low-frequency laser fields,” Phys. Rev. A 49, der, N. Milosevic, T. Brabec, P. B. Corkum, U. Heinz- 2117 (1994). mann, M. Drescher, and F. Krausz, “Attosecond [18] R. Pazourek, S. Nagele, and J. Burgdörfer,“Attosecond metrology,” Nature 414, 509–513 (2001). chronoscopy of photoemission,” Rev. Mod. Phys. 87, [4] A. Baltuska, Th. Udem, M. Uiberacker, M. Hentschel, 765 (2015). E. Goulielmakis, Ch. Gohle, R. Holzwarth, V. S. [19] C. J. Joachain, N. J. Kylstra, and R. M. Potvliege, Yakovlev, A. Scrinzi, T. W. Hansch, and F. Krausz, Atoms in Intense Laser Fields (Cambridge University “Attosecond control of electronic processes by intense Press, Cambridge, England, 2012). light fields,” Nature 421, 611–615 (2003). [20] A. L’Huiller, M. Lewenstein, P. Salières,Ph. Balcou, [5] P. Salières,A. L’Huillier, P. Antoine, and M. Lewen- M. Yu. Ivanov, J. Larsson, and C. G. Wahlström, stein, “Study of the spatial and temporal coherence of “High-order harmonic-generation cutoff,” Phys. Rev. A high-order harmonics,” in Advances in Atomic, Molecu- 48, R3433 (1993). lar and Optical Physics. Vol. 41, edited by B. Bederson [21] J. P. Marangos, “Development of high harmonic and H. Walter (Academic Press, San Diego, 1999) pp. generation spectroscopy of organic molecules and 83–142. biomolecules,” J. Phys. B 49, 132001 (2016). [6] D. Batani, C. J. Joachain, S. Martellucci, and A. N. [22] K. C. Kulander, K. J. Schafer, and J. L. Krause, Chester, Atoms, Solids, and Plasmas in Super-Intense “Dynamics of short-pulse excitation, ionization and Laser Fields (Kluwer Academic/Plenum, New York, harmonic conversion,” in Super-Intense Laser-Atom 2001). Physics, edited by B. Piraux, A. L’ Huillier, and [7] M. Lewenstein and A. L’Huillier, “Principles of single K. Rzazewski (Plenum, New York, 1993) pp. 95–110. atom physics: High-order harmonic generation, above- [23] H. B. van Linden van den Heuvell and H. G. Muller, in threshold ionization and non-sequential ionization,” Multiphoton Processes, edited by S. J. Smith and P. L. in Strong Field Laser Physics, edited by T. Brabec Knight (Cambridge University Press, Cambridge, Eng- (Springer, New York, 2009) pp. 147–183. land, 1988) p. 25. [8] P. B. Corkum and F. Krausz, “Attosecond science,” [24] M. Yu. Kuchiev, “Atomic antenna,” Sov. Phys. JETP Nat. Phys. 3, 381–387 (2007). 45, 404–406 (1987). [9] L. V. Keldysh, “Ionization in the field of a strong [25] T. Brabec and F. Krausz, “Intense few-cycle laser fields: electromagnetic wave,” J. Expt. Theo. Phys. 20, 1307 Frontiers of nonlinear optics,” Rev. Mod. Phys. 72, 545– (1965). 591 (2000). [10] A. M. Perelomov, V. S. Popov, and M. V. Terentev, [26] H. G. Muller, H. B. van Linden van den Heuvell, and “Ionization of atoms in an alternating electric field,” M. J. van der Wiel, “Experiments on ”above-threshold Sov. Phys. JETP 23, 924 (1966). ionization” of atomic hydrogen,” Phys. Rev. A 34, 236 [11] H. R. Reiss, “Effect of an intense electromagnetic field (1986). on a weakly bound system,” Phys. Rev. A 22, 1786 [27] P. Agostini, F. Fabre, G. Mainfray, G. Petite, and (1980). N. K. Rahman, “Free-free transitions following six- [12] M. V. Ammosov, N. B. Delone, and V. P. Krainov, photon ionization of xenon atoms,” Phys. Rev. Lett. “Tunnel ionization of complex atoms and of atomic 42, 1127 (1979). ions in an alternating electromagnetic field,” Sov. Phys. [28] D. B. Milo˘sević,G.G. Paulus, D. Bauer, and W. Becker, JETP 64, 1191 (1986). “Above-threshold ionization by few-cycle pulses,” J. [13] F. H. M. Faisal, Theory of Multiphoton processes Phys. B 39, R203–R262 (2006). (Springer, New York, 1987). [29] G. G. Paulus, W. Nicklich, X. Huale, P. Lambropoulus, [14] A. S. Landsman and U. Keller, “Attosecond science and and H. Walter, “Plateau in above threshold ionization the tunnelling time problem,” Phys. Rep. 547, 1–24 spectra,” Phys. Rev. Lett. 72, 2851 (1994). (2015). [30] G. G. Paulus, F. Lindner, H. Walther, A. Baltu˘ska, [15] P. B. Corkum, “Plasma perspective on strong field mul- E. Goulielmakis, M. Lezius, and F. Krausz, “Measure- tiphoton ionization,” Phys. Rev. Lett. 71, 1994 (1993). ment of the phase of few-cycle laser pulses,” Phys. Rev. 18

Lett. 91, 253004 (2003). [45] E. T. Gumbrell, A. J. Comley, M. H. R. Hutchinson, [31] W. Becker, F. Grasbon, R. Kopold, D. B. Milo˘sević, and R. A. Smith, “Intense laser interactions with sprays G. G. Paulus, and H. Walther, “Above-threshold ion- of submicron droplets,” Phys. Plasmas 8, 1329 (2001). ization: From classical features to quantum effects,” in [46] T. D. Donnelly, M. Rust, I. Weiner, M. Allen, R. A. Advances In Atomic, Molecular, and Optical Physics, Smith, C. A. Steinke, S. Wilks, J. Zweiback, T. E. edited by B. Bederson and H. Walter (Academic Press, Cowan, and T. Ditmire, “Hard x-ray and hot electron San Diego, 2002) p. 35. production from intense laser irradiation of wavelength- [32] P. Salières,B. Carré,L. Le Déroff,F. Grasbon, G. G. scale particles,” J. Phys. B 34, L313 (2001). Paulus, H. Walther, R. Kopold, W. Becker, D. B. Milo- [47] L. C. Mountford, R. A. Smith, and M. H. R. Hutchin- sevic, A. Sanpera, and M. Lewenstein, “Feynman’s son, “Characterization of a sub-micron liquid spray for path-integral approach for intense-laser-atom interac- laser-plasma x-ray generation,” Rev. Sci. Instrum. 69, tions,” Science 292, 902–905 (2001). 3780 (1998). [33] A. L’Huiller, L. A. Lompre, G. Mainfray, and C. Manus, [48] H. A. Sumeruk, S. Kneip, D. R. Symes, I. V. Chu- “Multiply charged ions induced by multiphoton absorp- rina, A. V. Belolipetski, G. Dyer, J. Landry, G. Bansal, tion in rate gases at 0.53 µm,” Phys. Rev. A 27, 2503 A. Bernstein, T. D. Donnelly, A. Karmakar, A. Pukhov, (1983). and T. Ditmire, “Hot electron and x-ray produc- [34] B. Walker, B. Sheehy, L. F. DiMauro, P. Agostini, K. J. tion from intense laser irradiation of wavelength-scale Schafer, and K. C. Kulander, “Precision measurement polystyrene spheres,” Phys. Plasmas 14, 062704 (2007). of strong field double ionization of helium,” Phys. Rev. [49] H. A. Sumeruk, S. Kneip, D. R. Symes, I. V. Churina, Lett. 73, 1227 (1994). A. V. Belolipetski, T. D. Donnelly, and T. Ditmire, [35] O. Smirnova, Y. Mairesse, S. Patchkovskii, N. Du- “Control of strong-laser-field coupling to electrons in dovich, D. Villeneuve, P. B. Corkum, and M. Yu. solid targets with wavelength-scale spheres,” Phys. Rev. Ivanov, “High harmonic interferometry of multi-electron Lett. 98, 045001 (2007). dynamics in molecules,” Nature 460, 972–977 (2009). [50] H. Li, B. Mignolet, G. Wachter, S. Skruszewicz, [36] A. D. Shiner, B. E. Schmidt, C. Trallero-Herrero, H. J. S. Zherebtsov, F. Süßmann, A. Kessel, S. A. Trushin, Wörner,S. Patchkovskii, P. B. Corkum, J-C. Kieffer, N. G. Kling, M. Kübel, B. Ahn, D. Kim, I. Ben- F. Légaré, and D. M. Villeneuve, “Probing collective Itzhak, C. L. Cocke, T. Fennel, J. Tiggesbäumker, K.- multi-electron dynamics in xenon with high-harmonic H. Meiwes-Broer, C. Lemell, J. Burgdörfer,R.D. Levine, spectroscopy,” Nat. Phys. 7, 464–467 (2011). F. Remacle, and M. F. Kling, “Coherent electronic wave [37] B. Bergues, M. Kübel, N. G. Kling, C. Burger, and packet motion in C60 controlled by the waveform and M. F. Kling, “Single-cycle non-sequential double ioniza- polarization of few-cycle laser fields,” Phys. Rev. Lett. tion,” IEEE J. of Sel. Top. in Quant. Elect. 21, 8701009 114, 123004 (2015). (2015). [51] V. Yakovlev, M. I. Stockman, F. Krausz, and P. Baum, [38] Y. L. Shao, T. Ditmire, J. W. G. Tisch, E. Springate, “Atomic-scale diffractive imaging of sub-cycle electron J. P. Marangos, and M. H. R. Hutchinson, “Multi- dynamics in condensed matter,” Sci. Rep. 5, 14581 kev electron generation in the interaction of intense (2015). laser pulses with xe clusters,” Phys. Rev. Lett. 77, 3343 [52] M. Kübel, R. Siemering, C. Burger, N. G. Kling, H. Li, (1996). A.S. Alnaser, B. Bergues, S. Zherebtsov, A. M. Azzeer, [39] T. Ditmire, J. W. G. Tisch, E. Springate, M. B. Ma- I. Ben-Itzhak, R. Moshammer, R. de Vivie-Riedle, and son, N. Hay, R. A. Smith, J. Marangos, and M. H. R. M. F. Kling, “Steering proton migration in hydrocar- Hutchinson, “High-energy ions produced in explosions bons using intense few-cycle laser fields,” Phys. Rev. of superheated atomic clusters,” Nature (London) 386, Lett. 116, 193001 (2016). 54 (1997). [53] L. Belshaw, F. Calegari, M.J. Duffy, A. Trabattoni, [40] T. Ditmire, R. A. Smith, J. W. G. Tisch, and M. H. R. L. Poletto, M. Nisoli, and J.B. Greenwood, “Obser- Hutchinson, “High intensity laser absorption by gases vation of ultrafast charge migration in an amino acid,” of atomic clusters,” Phys. Rev. Lett. 78, 3121 (1997). J. Phys. Chem. Lett. 3, 3751–3754 (2012). [41] R. A. Smith, T. Ditmire, and J. W. G. Tisch, “Char- [54] F. Calegari, D. Ayuso, A. Trabattoni, L. Belshaw, S. De acterization of a cryogenically cooled high-pressure gas Camillis, S. Anumula, F. Frassetto, L. Poletto, A. Pala- jet for laser/cluster interaction experiments,” Rev. Sci. cios, P. Decleva, J. B. Greenwood, F. Mart´ın, and Instrum. 69, 3798 (1998). M. Nisoli, “Ultrafast electron dynamics in phenylala- [42] J. W. G. Tisch, T. Ditmire, D. J. Fraser, N. Hay, M. B. nine initiated by attosecond pulses,” Science 336, 346 Mason, E. Springate, J. P. Marangos, and M. H. R. (2014). Hutchinson, “Investigation of high-harmonic generation [55] C. I. Blaga, J. Xu, A. D. DiChiara, E. Sistrunk, from xenon atom clusters,” J. Phys. B 30, L709 (1997). K. Zhang, P. Agostini, T. A. Miller, L. F. DiMauro, and [43] E. Skopalová,Y. C. El-Taha, A. Za¨ır,M. Hohenberger, C. D. Lin, “Imaging ultrafast molecular dynamics with E. Springate, J. W. G. Tisch, R. A. Smith, and J. P. laser-induced electron diffraction,” Nature 483, 194–197 Marangos, “Pulse-length dependence of the anisotropy (2012). of laser-driven cluster explosions: Transition to the [56] J. Xu, C. I. Blaga, K. Zhang, Y. H. Lai, C. D. Lin, T. A. impulsive regime for pulses approaching the few-cycle Miller, P. Agostini, and L. F. DiMauro, “Diffraction us- limit,” Phys. Rev. Lett. 104, 203401 (2010). ing laser-driven broadband electron wave packets,” Nat. [44] D. R. Symes, A. J. Comley, and R. A. Smith, “Fast- Comm. 5, 4635 (2014). ion production from short-pulse irradiation of ethanol [57] M. G. Pullen, B. Wolter, A-T. Le, M. Baudisch, microdroplets,” Phys. Rev. Lett. 93, 145004 (2004). M. Hemmer, A. Senftleben, C. D. Schröter,J. Ullrich, R. Moshammer, C. D. Lin, and J. Biegert, “Imaging an 19

aligned polyatomic molecule with laser-induced electron makis, F. Krausz, and V. S. Yakovlev, “Delay in pho- diffraction,” Nat. Comm. 6, 7262 (2015). toemission,” Science 328, 1658–1662 (2010). [58] P. Hommelhoff and M. F. Kling, Attosecond [71] S. Neppl, R. Ernstorfer, E. M. Bothschafter, A. L. Cava- Nanophysics: From Basic Science to Applications lieri, D. Menzel, J. V. Barth, F. Krausz, R. Kienberger, (Wiley-VCH, Berlin, 2015). and P. Feulner, “Attosecond time-resolved photoemis- [59] A. Schiffrin, T. Paasch-Colberg, N. Karpowicz, sion from core and valence states of magnesium,” Phys. V. Apalkov, D. Gerster, S. Mühlbrandt, M. Korb- Rev. Lett. 109, 087401 (2012). man, J. Reichert, M. Schultze, S. Holzner, J. V. Barth, [72] A. L. Cavalieri, N. Müller,Th. Uphues, V. S. Yakovlev, R. Kienberger, R. Ernstorfer, V. S. Yakovlev, M. I. A. Baltuˇska, B. Horvath, B. Schmidt, L. Blümel, Stockman, and F. Krausz, “Optical-field-induced cur- R. Holzwarth, S. Hendel, M. Drescher, U. Kleineberg, rent in dielectrics,” Nature 493, 70–74 (2013). P. M. Echenique, R. Kienberger, F. Krausz, and [60] M. Schultze, E. M Bothschafter, A. Sommer, S. Holzner, U Heinzmann, “Attosecond spectroscopy in condensed W. Schweinberger, M. Fiess, M. Hofstetter, R. Kien- matter,” Nature 449, 1029–1032 (2007). berger, V. Apalkov, V. S. Yakovlev, M. I. Stockman, [73] R. Locher, L. Castiglioni, M. Lucchini, M. Greif, and F. Krausz, “Controlling dielectrics with the electric L. Gallmann, J. Osterwalder, M. Hengsberger, and field of light,” Nature 493, 75–78 (2013). U. Keller, “Energy-dependent photoemission delays [61] F. Süßmann, L. Seiffert, S. Zherebtsov, V. Mon- from noble metal surfaces by attosecond interferome- des, J. Stierle, M. Arbeiter, J. Plenge, P. Rupp, try,” Optica 2, 405 (2015). C. Peltz, A. Kessel, S. A. Trushin, B. Ahn, D. Kim, [74] W. A. Okell, T. Witting, D. Fabris, C. A. Arrell, C. Graf, E. Rühl,M. F. Kling, and T. Fennel, “Field J. Hengster, S. Ibrahimkutty, A. Seiler, M. Barthelmess, propagation-induced directionality of carrier-envelope S. Stankov, D. Y. Lei, Y. Sonnefraud, M. Rahmani, phase-controlled photoemission from nanospheres,” T. Uphues, S. A. Maier, J. P. Marangos, and J. W. G. Nat. Comm. 6, 7944 (2015). Tisch, “Temporal broadening of attosecond photoelec- [62] S. Zherebtsov, T. Fennel, J. Plenge, E. Antonsson, tron wavepackets from solid surfaces,” Optica 2, 383– I. Znakovskaya, A. Wirth, O. Herrwerth, F. Süßmann, 387 (2015). C. Peltz, I. Ahmad, S. A. Trushin, V. Pervak, S. Karsch, [75] M. I Stockman, “Nanoplasmonics: past, present, and M. J. J. Vrakking, B. Langer, C. Graf, M. I. Stock- glimpse into future,” Opt. Exp. 19, 22029–22106 (2011). man, F. Krausz, E. Rühl,and M. F. Kling, “Controlled [76] S. Kim, J. Jin, Y-J. Kim, I-Y. Park, Y. Kim, and S-W. near-field enhanced electron acceleration from dielectric Kim, “High-harmonic generation by resonant plasmon nanospheres with intense few-cycle laser fields,” Nat. field enhancement,” Nature 453, 757–760 (2008). Phys. 7, 656–662 (2011). [77] M. Sivis, M. Duwe, B. Abel, and C. Ropers, “Extreme- [63] M. Krüger,M. Schenk, and P. Hommelhoff, “Attosec- ultraviolet light generation in plasmonic nanostruc- ond control of electrons emitted from a nanoscale metal tures,” Nat. Phys. 9, 304–309 (2013). tip,” Nature 475, 78–81 (2011). [78] I-Y. Park, S. Kim, J. Choi, D-H. Lee, Y. J. Kim, M. F. [64] G. Herink, D. R. Solli, M. Gulde, and C. Ropers, “Field- Kling, M. I. Stockman, and S-W. Kim, “Plasmonic gen- driven photoemission from nanostructures quenches the eration of ultrashort extreme-ultraviolet light pulses,” quiver motion,” Nature 483, 190–193 (2012). Nat. Phot. 5, 677 (2011). [65] B. Piglosiewicz, S. Schmidt, D. J. Park, J. Vogel- [79] N. Pfullmann, C. Waltermann, M. Noack, S. Rausch, sang, P. Gross, C. Manzoni, P. Farinello, G. Cerullo, T. Nagy, C. Reinhardt, M. Kova˘cev, V. Knittel, and C. Lienau, “Carrier-envelope phase effects on the R. Bratschitsch, D. Akemeier, A. Hütten,A. Leiten- strong-field photoemission of electrons from metallic storfer, and U. Morgner, “Bow-tie nano-antenna as- nanostructures,” Nat. Phot. 8, 37–42 (2014). sisted generation of extreme ultraviolet radiation,” New [66] S. Ghimire, A. D. DiChiara, E. Sistrunk, P. Agostini, J. Phys. 15, 093027 (2013). L. F. DiMauro, and D. A. Reis, “Observation of high- [80] M. Sivis, M. Duwe, B. Abel, and C. Ropers, order harmonic generation in a bulk crystal,” Nat. Phys. “Nanostructure-enhanced atomic line emission,” Nature 7, 138–141 (2011). 485, E1–E3 (2012). [67] T. T. Luu, M. Garg, S. Yu. Kruchinin, A. Moulet, [81] S. Kim, J. Jin, Y-J. Kim, I-Y. Park, Y. Kim, and S-W. M. Th. Hassan, and E. Goulielmakis, “Extreme ultravi- Kim, “Reply nature),” Nature 485, E1–E3 (2012). olet high-harmonic spectroscopy of solids,” Nature 521, [82] D. J. Park, B. Piglosiewicz, S. Schmidt, H. Kollmann, 498–502 (2015). M. Mascheck, P. Groß, and C. Lienau, “Character- [68] O. Schubert, M. Hohenleutner, F. Langer, B. Urbanek, izing the optical near-field in the vicinity of a sharp C. Lange, U. Huttner, D. Golde, T. Meier, M. Kira, metallic nanoprobe by angle-resolved electron kinetic S. W. Koch, and R. R. Huber, “Sub-cycle control of energy spectroscopy,” Ann. Phys. (Berlin) 525, 135–142 terahertz high-harmonic generation by dynamical bloch (2013). oscillations,” Nat. Phot. 8, 119–123 (2014). [83] M. F. Ciappina, J. Biegert, R. Quidant, and M. Lewen- [69] G. Vampa, T. J. Hammond, N. Thiré,B. E. Schmidt, stein, “High-order-harmonic generation from inhomoge- F. Legaré, C. R. McDonald, T. Brabec, and P. B. neous fields,” Phys. Rev. A 85, 033828 (2012). Corkum, “Linking high harmonics from gases and [84] M. F. Ciappina, S. S. Aćimović,T. Shaaran, J. Biegert, solids,” Nature 522, 462–464 (2015). R. Quidant, and M. Lewenstein, “Enhancement of high [70] M. Schultze, M. Fieß, N. Karpowicz, J. Gagnon, harmonic generation by confining electron motion in M. Korbman, M. Hofstetter, S. Neppl, A. L. Cavalieri, plasmonic nanostrutures,” Opt. Exp. 20, 26261–26274 Y. Komninos, Th. Mercouris, C. A. Nicolaides, R. Pa- (2012). zourek, S. Nagele, J. Feist, J. Burgdörfer,A. M. Azzeer, [85] A. Husakou, S-J. Im, and J. Herrmann, “Theory of R. Ernstorfer, R. Kienberger, U. Kleineberg, E. Gouliel- plasmon-enhanced high-order harmonic generation in 20

the vicinity of metal nanostructures in noble gases,” order harmonic generation: the role of the field inho- Phys. Rev. A 83, 043839 (2011). mogeneity,” J. Mod. Opt. 86, 1634–1639 (2012). [86] I. Yavuz, E. A. Bleda, Z. Altun, and T. Topcu, “Gen- [102] I. Yavuz, “Gas population effects in harmonic emission eration of a broadband XUV continuum in high-order- by plasmonic fields,” Phys. Rev. A 87, 053815 (2013). harmonic generation by spatially inhomogeneous fields,” [103] M. F. Ciappina, T. Shaaran, and M. Lewenstein, “High Phys. Rev. A 85, 013416 (2012). order harmonic generation in noble gases using plas- [87] E. Neyra, F. Videla, M. F. Ciappina, J. A. Perez- monic field enhancement,” Ann. Phys. (Berlin) 525, 97– Hernandez, L. Roso, M. Lewenstein, and G. A. Torchia, 106 (2013). “High-order harmonic generation driven by inhomoge- [104] M. F. Ciappina, T. Shaaran, R. Guichard, J. A. Pérez- neous plasmonics fields spatially bounded: influence on Hernández, L. Roso, M. Arnold, T. Siegel, A. Za¨ır, the cut-off law,” J.Opt. 20, 034002 (2018). and M. Lewenstein, “High energy photoelectron emis- [88] T. Shaaran, M. F. Ciappina, R. Guichard, J. A. Pérez- sion from gases using plasmonic enhanced near-fields,” Hernández,L. Roso, M. Arnold, T. Siegel, A. Za¨ır, and Las. Phys. Lett. 10, 105302 (2013). M. Lewenstein, “High-order-harmonic generation by en- [105] M. F. Ciappina, J. A. Pérez-Hernández,T. Shaaran, hanced plasmonic near-fields in metal nanoparticles,” M. Lewenstein, M. Krüger,and P. Hommelhoff, “High- Phys. Rev. A 87, 041402(R) (2013). order harmonic generation driven by metal nanotip pho- [89] K. J. Schafer, “The energy analysis of time-dependent toemission: theory and simulations,” Phys. Rev. A 89, numerical wave functions,” Comp. Phys. Comm. 63, 013409 (2014). 427–434 (1991). [106] M. F. Ciappina, J. A. Pérez-Hernández,T. Shaaran, [90] K. J. Schafer and K. C. Kulander, “Energy analysis of and M. Lewenstein, “Coherent xuv generation driven time-dependent wave functions: Application to above- by sharp metal tips photoemission,” Eur. Phys. J. D threshold ionization,” Phys. Rev. A 42, 5794(R) (1990). 68, 172 (2014). [91] K. J. Schafer, “Numerical methods in strong field [107] M. F. Ciappina, J. A. Pérez-Hernández, L. Roso, physics,” in Strong Field Laser Physics, edited by A. Za¨ır, , and M. Lewenstein, “High-order harmonic T. Brabec (Springer, New York, 2009) pp. 111–145. generation driven by plasmonic fields: a new route to- [92] J. A. Pérez-Hernández,M. F. Ciappina, M. Lewenstein, wards the generation of uv and xuv photons?” J. Phys.: L. Roso, and A. Za¨ır,“Beyond Carbon K-Edge Har- Conf. Ser. 601, 012001 (2015). monic Emission Using a Spatial and Temporal Synthe- [108] A. Husakou and J. Herrmann, “Quasi-phase-matched sized Laser Field,” Phys. Rev. Lett. 110, 053001 (2013). high-harmonic generation in composites of metal [93] M. F. Ciappina, J. A. Pérez-Hernández,T. Shaaran, nanoparticles and a noble gas,” Phys. Rev. A 90, 023831 L. Roso, and M. Lewenstein, “Electron-momentum dis- (2014). tributions and photoelectron spectra of atoms driven by [109] H. Ebadi, “Interferences induced by spatially nonhomo- an intense spatially inhomogeneous field,” Phys. Rev. A geneous fields in high-harmonic generation,” Phys. Rev. 87, 063833 (2013). A 89, 053413 (2014). [94] T. Shaaran, M. F. Ciappina, and M. Lewenstein, [110] B. Fetić,K. Kalajd˘zić, and D. B. Milo˘sević,“High- “Quantum-orbit analysis of above-threshold ionization order harmonic generation by a spatially inhomogeneous driven by an intense spatially inhomogeneous field,” field,” Ann. Phys. (Berlin) 525, 107–117 (2012). Phys. Rev. A 87, 053415 (2013). [111] J. Luo, Y. Li, Z. Wang, Q. Zhang, and P. Lu, “Ultra- [95] M. F. Ciappina, J. A. Pérez-Hernández,and M. Lewen- short isolated attosecond emission in mid-infrared inho- stein, “Classstrong: Classical simulations of strong field mogeneous fields without cep stabilization,” J. Phys. B processes,” Comp. Phys. Comm. 185, 398–406 (2014). 46, 145602 (2013). [96] A. Chacón,M. F. Ciappina, and M. Lewenstein, “Nu- [112] L. Feng, M. Yuan, and T. Chu, “Attosecond x-ray merical studies of light-matter interaction driven by source generation from two-color polarized gating plas- plasmonic fields: The velocity gauge,” Phys. Rev. A 92, monic field enhancement,” Phys. Plasmas 20, 122307 063834 (2015). (2013). [97] I. Yavuz, Y. Tikman, and Z. Altun, “High-order- [113] Z. Wang, P. Lan, J. Luo, L. He, Q. Zhang, and P. Lu, + harmonic generation from H2 molecular ions near “Control of electron dynamics with a multicycle two- plasmon-enhanced laser fields,” Phys. Rev. A 92, color spatially inhomogeneous field for efficient single- 023413 (2015). attosecond-pulse generation,” Phys. Rev. A 88, 063838 [98] A. Husakou, F. Kelkensberg, J. Herrmann, and M. J. J. (2013). Vrakking, “Polarization gating and circularly-polarized [114] J. Luo, Y. Li, Z. Wang, L. He, Q. Zhang, and P. Lu, “Ef- high harmonic generation using plasmonic enhancement ficient supercontinuum generation by uv-assisted mid- in metal nanostructures,” Opt. Exp. 19, 25346–25354 infrared plasmonic fields,” Phys. Rev. A 89, 023405 (2011). (2013). [99] T. Shaaran, M. F. Ciappina, and M. Lewenstein, [115] L. He, Z. Wang, Y. Li, Q. Zhang, P. Lan, and P. Lu, “Quantum-orbit analysis of high-order-harmonic gener- “Wavelength dependence of high-order-harmonic yield ation by resonant plasmon field enhancement,” Phys. in inhomogeneous fields,” Phys. Rev. A 88, 053404 Rev. A 86, 023408 (2012). (2013). [100] M. F. Ciappina, J. A. Pérez-Hernández,T. Shaaran, [116] C. Zang, C. Lui, and Z. Xu, “Control of higher spectral J. Biegert, R. Quidant, and M. Lewenstein, “Above- components by spatially inhomogeneous fields in quan- threshold ionization by few-cycle spatially inhomoge- tum wells,” Phys. Rev. A 88, 035805 (2013). neous fields,” Phys. Rev. A 86, 023413 (2012). [117] J. Luo, Y. Li, Z. Wang, Q. Zhang, P. Lan, and P. Lu, [101] T. Shaaran, M. F. Ciappina, and M. Lewenstein, “Es- “Wavelength dependence of high-order-harmonic yield timating the plasmonic field enhancement using high- in inhomogeneous fields,” J. Opt. Soc. Am. B 30, 2469– 21

2475 (2013). tructures and nanoparticles,” in Progress in Nonlin- [118] X. Cao, S. Jiang, C. Yu, Y. Wang, L. Bai, and R. Lu, ear Nano-Optics, edited by S. Sakabe, C. Lienau, and “Generation of isolated sub-10-attosecond pulses in spa- R. Grunwald (Springer International Publishing, 2015) tially inhomogenous two-color fields,” Opt. Exp. 22, pp. 251–268. 26153–26161 (2014). [131] A. M. Sayler, T. Rathje, W. Müller,K. Rühle,R. Kien- [119] Z. Wang, L. He, J. Luo, P. Lan, and P. Lu, “High-order berger, and G. G. Paulus, “Precise, real-time, every- harmonic generation from rydberg atoms in inhomoge- single-shot, carrier-envelope phase measurement of ul- neous fields,” Opt. Exp. 22, 25909–25922 (2014). trashort laser pulses,” Opt. Lett 36, 1–3 (2011). [120] L. Feng and H. Liu, “Attosecond extreme ultraviolet [132] G. G. Paulus, F. Grasbon, H. Walther, P. Villoresi, generation in cluster by using spatially inhomogeneous M. Nisoli, S. Stagira, E. Priori, and S. De Silvestri, field,” Phys. Plasmas 22, 013107 (2015). “Absolute-phase phenomena in photoionization with [121] C. Yu, Y. Wang, X. Cao, S. Jiang, and R. Lu, “Isolated few-cycle laser pulses,” Nature 414, 182–184 (2001). few-attosecond emission in a multi-cycle asymmetrically [133] M. Krüger, M. Förster, and P. Hommelhoff, “Self- nonhomogeneous two-color laser field,” J. Phys. B 47, probing of metal nanotips by rescattered electrons re- 225602 (2015). veals the nano-optical near-field,” J. Phys. B 47, 124022 [122] C. Zagoya, M. Bonner, H. Chomet, E. Slade, and (2014). C. Figueira de Morisson Faria, “Different time scales [134] A. Apolonski, P. Dombi, G. G. Paulus, M. Kakehata, in plasmonically enhanced high-order-harmonic genera- R. Holzwarth, Th. Udem, Ch. Lemell, K. Torizuka, tion,” Phys. Rev. A 93, 053419 (2016). J. Burgdörfer,T. W. Hänsch, and F. Krausz, “Ob- [123] J. A. Pérez-Hernández, D. J. Hoffmann, A. Za¨ır,L. E. servation of light-phase-sensitive photoemission from a Chipperfield, L. Plaja, C. Ruiz, J. P. Marangos, and metal,” Phys. Rev. Lett. 92, 073902 (2004). L. Roso, “Extension of the cut-off in high-harmonic gen- [135] D. G. Arbó,J. E. Miraglia, M. S. Gravielle, K. Schiessl, eration using two delayed pulses of the same colour,” J. E. Persson, and J. Burgdörfer,“Coulomb-volkov ap- Phys. B 42, 134004 (2009). proximation for near-threshold ionization by short laser [124] F. Süßmannand M. F. Kling, “Attosecond measurement pulses,” Phys. Rev. A 77, 013401 (2008). of petahertz plasmonic near-fields,” Proc. of SPIE 8096, [136] L. Ortmann, J. A. Pérez-Hernández,M. F. Ciappina, 80961C (2011). J. Schötz,A. Chacón,G. Zeraouli, M. F. Kling, L. Roso, [125] F. Süßmannand M. F. Kling, “Attosecond nanoplas- M. Lewenstein, and A. S. Landsman, “Emergence of a monic streaking of localized fields near metal higher energy structure in strong field ionization with nanospheres,” Phys. Rev. B 84, 121406(R) (2011). inhomogeneous electric fields,” Phys. Rev. Lett. 119, [126] P. Hommelhoff, Y. Sortais, A. Aghajani-Talesh, and 053204 (2017). M. A. Kasevich, “Field emission tip as a nanometer [137] I. Yavuz, M. F. Ciappina, A. Chacón,Z. Altun, M. F. source of free electron femtosecond pulses,” Phys. Rev. Kling, and M. Lewenstein, “Controlling electron local- + Lett. 96, 077401 (2006). ization in H2 by intense plasmon-enhanced laser fields,” [127] M. Schenk, M. Krüger, and P. Hommelhoff, “Strong- Phys. Rev. A 93, 033404 (2016). field above-threshold photoemission from sharp metal [138] A. Chacón,M. F. Ciappina, and M. Lewenstein, “Sig- tips,” Phys. Rev. Lett. 105, 257601 (2010). natures of double-electron recombination in high-order [128] M. Krüger,M. Schenk, P. Hommelhoff, G. Wachter, harmonic generation driven by spatial inhomogeneous C. Lemell, and J. Burgdörfer,“Interaction of ultra- fields,” (2015), submitted. short laser pulses with metal nanotips: a model system [139] Y. Tikman, I. Yavuz, M. F. Ciappina, A. Chacón,Z. Al- for strong-field phenomena,” New J. Phys. 14, 085019 tun, and M. Lewenstein, “High-order-harmonic gener- (2012). ation from rydberg atoms driven by plasmon-enhanced [129] M. Krüger,M. Schenk, M. Förster,and P. Hommelhoff, laser fields,” Phys. Rev. A 93, 023410 (2016). “Attosecond physics in photoemission from a metal nan- [140] S. Han, H. Kim, Y. W. Kim, Y.-J. Kim, S. Kim, I.- otip,” J. Phys. B 45, 074006 (2012). Y. Park, and S.-W. Kim, “High harmonic genera- [130] A. Husakou, S-J. Im, K.H. Kim, and J. Herrmann, tion by strongly enhanced femtosecond pulses in metal- “High harmonic generation assisted by metal nanos- sapphire nanostructure waveguide,” Nat. Commun. 7, 13105 (2016).